Cahiers d'études

We analyze the economic impact of the green transition in the euro area by extending the Euro Area and Global Economy (EAGLE) model to include green and brown energy sectors. In this model, energy goods are consumed both as final goods by households and as inputs by intermediate goods firms. A carbon tax acts as a cost-push shock, creating stagflation-ary effects, particularly when fiscal interventions are not primary-balance neutral. Without subsidies for green energy firms, the green transition is limited to a shift in household ex-penditure towards green energy goods. However, when authorities provide subsidies to green energy firms, intermediate goods firms also increase their demand for green energy inputs, thereby strengthening the demand channel in the green energy market and driving its price upward. When carbon taxes are raised globally and governments redistribute carbon tax revenues to green energy firms, the recession in the euro area deepens, and inflationary pres-sures increase, partly due to a weakening euro. Taxes on brown capital investment are also contractionary but lead to lower inflation. In this scenario, subsidies for investment in green energy capital can help mitigate the recession. However, overall, taxes on brown capital investment are less effective in driving the green transition compared to carbon taxes.

JEL Classification: C53, E32, E52, F45, H30, Q48.
Keywords: Climate Policy, Carbon Taxation, Monetary Policy, Fiscal Policy, Euro Area,
DSGE modeling

We study the effects on capital accumulation from public health in an overlapping generations model. Investing in the health of young individuals raises longevity and lowers frailty, influencing capital accumulation through three main channels. First, since it is tax-financed, public health investment reduces disposable income and the capacity to save (cost channel).
Second, it prolongs life expectancy, encouraging individuals to save for old age (longevity channel). Third, it reduces frailty and the need to save to finance long-term care (frailty channel). Longevity and frailty have ambiguous effects on taxation when the government subsidizes long-term care. We analytically derive the economic implications of these health channels and numerically illustrate our findings. Our main result is that although public investment in healthy aging is costly, it can stimulate capital accumulation even without directly affecting productivity.

Keywords: Overlapping generations, public health, health channels, capital accumulation.
JEL-Code: H51, I15, O41

Abstract:
This report presents the methodology and main descriptive results of the fourth wave of the Cross-border Household Finance and Consumption Survey (XB-HFCS) conducted in 2021. This is a household survey of employees in Luxembourg who live abroad and regularly commute across the border. We analyse the composition and level of their assets and liabilities, net wealth and income, and compare them to similar households (including at least one employee) that live and work in Luxembourg, Belgium, France or Germany. Compared to households who were employed in their country of residence, cross-border commuters were more likely to be homeowners and tended to enjoy higher gross income and net wealth.
Keywords: Cross-border commuters, households, survey, assets, liabilities, wealth, income. JEL-Codes: G51, D31, D14, C81, C83, J61

Abstract. To finance the fight against climate change, sustainable investment is projected to surpass $40 trillion by 2030. In principle, sustainable investment diverts funds away from brown firms, increasing their borrowing costs to encourage them to become greener. However, recent empirical evidence does not support this channel, as the most polluting firms tend to become more brown in response to higher costs of capital. I formalise this empirical finding by developing a stylised model where brown firms must choose the optimal time to switch from old, polluting technologies to new, clean alternatives. Results indicate that raising the cost of capital for brown firms can have non-monotonic effects on the optimal switching times. For example, firms operating in capital-intensive sectors, often among the largest polluters, are more likely to respond to higher borrowing costs by delaying their switching time. In contrast, brown firms that are nearly ready to switch to cleaner methods may speed up their transition when faced with higher borrowing costs.
JEL Codes: Q50, Q56.

We propose a two-period overlapping generation economy that incorporates health investment in preventive measures during youth. These preventive measures contribute to increased longevity and reduced frailty, which influence old-age care costs. As these costs are funded through pay-as-you-go social security contributions, investment in prevention creates externalities for the next generation. We analytically determine the optimal level of prevention and characterize the optimal health policy that a government should implement to achieve it. Our findings reveal that the optimal subsidy to healthcare exceeds the optimal subsidy to preventive measures. Furthermore, both subsidies are inversely related to the generosity of the public pension scheme. We explore the robustness of our results through various extensions and demonstrate their consistency with several patterns observed in cross-country OECD data.
JEL Codes: H23, I18, O41.
Keywords: Health, Prevention, Optimal Ramsey policy, Overlapping generations.

This paper reviews the literature examining the consequences of heterogeneity in macroeconomic modeling, especially within the context of monetary and fiscal policy transmission. This review reveals that heterogeneity can significantly alter the transmission mechanisms of monetary policy in macroeconomic models and suggests possible advantages from collaboration between fiscal and monetary policies. The paper also provides a critical evaluation of various analytical and numerical methods to solve macroeconomic models with heterogeneity, underscoring the need for a careful methodological choice based on specific circumstances.
Keywords: Literature review, Heterogeneity, Monetary Policy, Fiscal Policy, Macroeconomic Modeling, HANK model.
JEL Codes: E32, E52, E62, D31, C61.

This technical report provides an introduction to solving economic models us-ing deep learning techniques. We offer a simple yet rigorous overview of deep learning methods and their applicability to economic modeling. We illustrate these concepts using the benchmark of modern macroeconomic theory: the stochastic growth model. Our results emphasize how various choices related to the design of the deep learning solution affect the accuracy of the results, providing some guidance for potential users of the method. We also provide fully commented computer codes. Overall, our hope is that this report will serve as an accessible, useful entry point to applying deep learning techniques to solve economic models for graduate students and researchers interested in the field.
JEL Codes: C45, C60, C63, E13.
Keywords: solutions of DSGE models; deep learning; artificial neural networks.

In this paper, we model two-way migration as the outcome of strategic public policies adopted by competing jurisdictions. We assume that two economies, distinguished by different technological levels, host a continuum of mobile individuals with varying skill levels. To maximize their net revenues, governments compete for mobile workers by taxing wages and providing a public good that enhances firm productivity (public input). We show that the most skilled workers migrate to the technologically advanced economy. However, the government in the less technologically developed economy can retain some of its skilled workers and attract workers from abroad by offering lower taxes or more public inputs. As a result, a two-way migration pattern emerges, driven by governments’ strategic policy choices. Finally, the introduction of heterogeneity in population size does not significantly alter the results.
Keywords: Bilateral migration; Tax competition; Heterogeneous skills; Technological gap; Policy competition

JEL Classifications: H20, H32, H54, H87, F22, F60

This paper introduces a forward-looking bank-level stress testing framework for a large-scale system to assess three forms of banking system vulnerability– bank capital fragility, bank capital adequacy and bank solvency. Results for Luxembourg are provided with a decomposition by bank business model and domicile type. The paper goes on to assess how these systemic risk indicators are linked to macroeconomic variables, and investigates their predictive power for Luxembourg’s nominal GDP growth one year ahead. Several important findings are documented over 2003Q2 to 2023Q3. First, the systemic risk indicators responded to the main stock market crashes in a timely manner. However, contributions from different bank business models and domicile types varied over time. Second, association with key macroeconomic variables (interest rates, liquidity flow, euro area consumer confidence and business climate) depended on the different characteristics of systemic risk across bank business models. Third, the systemic risk indicators contributed to explaining nominal GDP growth one year ahead. However, the systemic risk component associated with search-for-yield behavior and fee & commission generating activities could also explain nominal GDP growth, suggesting that if banks became more dependent on these income sources, they could create financial stability issues in the long run. Overall, the framework provides a useful monitoring toolkit that tracks changes in forward-looking systemic risk and risk spillovers in the Luxembourg banking sector.

JEL Classification: C1, E5, F3, G1
Keywords: financial stability; systemic risk; macro-prudential policy; dynamic dependence; banking business model; financial stress index; coronavirus Covid-19; macro-financial linkages.

This paper presents a new database of direct investment positions held through captive financial institutions (CFIs) in Luxembourg that are owned by (resident and non-resident) investment funds focusing on private equity or real estate. Compared to Di Filippo (2023), the new database is more comprehensive as it includes smaller CFIs with less than 500 million euros in total assets. Over 2011-2021, intra- Luxembourg investment positions were larger than foreign direct investment (FDI) positions (both inward and outward). Most of these intra-Luxembourg investment positions arise because investment funds use Luxembourg CFIs to structure their holdings and acquisitions across the globe. In 2021, only about 1% of the inward FDI position held through these CFIs is ultimately invested in targets located in Luxembourg. The outward FDI position held through these CFIs is invested in private companies (65%) or real estate assets (35%). While most investment fund sponsors are headquartered in the United States or in the United Kingdom, investment targets are mostly in Western Europe, with a focus on the euro area. Outward FDI in private equity targets companies that are quite dispersed across economic activities. In 2021, “Information, telecommunications and computer services” have the largest share (17%), followed by “Electricity, gas, water supply, recycling” (12%) and “Chemicals and non-metallic mineral products” (10%). Outward FDI in real estate is more concentrated by property type, with 45% in commercial buildings (office and retail properties), 24% in industrial buildings (in particular logistics properties) and 15% in residential properties. Targets in Luxembourg are mostly companies operating in finance and insurance activities (private equity) or office properties (real estate).

Keywords: Foreign direct investment, Captive financial institutions and money lenders, Sector S127, Investment funds, Private equity, Real estate 

JEL codes: C80, C81, F23, F30, G23, G32

Abstract. I explore the optimal timing of environmental policy when the stock of natural capital is unobserved and can only be imperfectly measured. I present two key insights. First, noisy signals about the natural capital stock blur the inference process, thereby easing the conditions under which policy adoption becomes optimal. Second, the interaction be tween natural capital stock volatility and the inference process gives rise to new effects that
are absent under perfect information. Specifically, the impact of increased volatility on the conditions for optimal policy adoption varies depending on the information set. My work contributes to both the environmental policy timing literature and the field of resource man agement under incomplete information.
JEL Codes: E20, Q50.
Keywords: Natural capital, Partial information, Optimal stopping.

This paper introduces new empirical findings concerning the rental housing market in the Paris metropolitan area. Combining a new dataset gathered from online advertisements for Parisian rentals with a hedonic model that incorporates both apartment features and property-specific photographs, two main stylized facts are established. First, with comparable property features, landlords who ask for lower rent attract a greater number of applicants, consistent with predictions from standard directed search models. Second, many landlords employ a two-stage pricing approach, initially advertising a high rent and then reducing it after a “wait-and-see” period. This previously unreported feature is consistent with the slow Dutch auction mecha-nism studied in the auction literature and observed in the property sales market.

Keywords: Rental Housing Market; Hedonic Model; Directed Search Models; Landlords’ Pricing Strategies; Machine Learning

Abstract. We build a life cycle model to study the implications of two types of lifetime uncertainty on investment in health and welfare. We show that when the hazard rate of death depends on age, uncertainty increases health investment. Instead, when hazard rate depends on human frailty, uncertainty decreases health investment. In both cases, uncertainty reduces welfare. The size of the effects depends on an aggregate parameter related to the natural increase in human frailty with age, to the marginal return on health investment and to the rate of time preference. We first derive the main results from a small model which admits an analytical solution, before generalizing them in a larger model using numerical simulations.
JEL Codes: C60, D15, D81, I12, I18.
Keywords: life cycle, uncertainty, health, welfare.

Abstract. This paper establishes that frictionless, rational-expectations models driven by specific ARMA(2,1) forcing processes are consistent with equilibrium asset-price dynamics featuring momentum. To reach this result, we first document that AR(2) models ade-quately capture the cyclical dynamics found in U.S. house prices, in particular the strong positive first-order autocorrelation in their first difference. Then, we show analytically that ARMA(2,1) exogenous drivers give rise to equilibrium AR(2) asset-price dynamics in a sim-ple present-value model. Our pen-and-paper approach yields a straightforward economic interpretation of the results, emphasizing the contribution of anticipated shocks to generat-ing asset-price momentum. We document the empirical relevance of our theoretical results by estimating the model from house-price data. Our findings suggest that house-price mo-mentum does not necessarily signal irrational exuberance or strong frictions in housing markets.
JEL Codes: C32, E32, G12.
Keywords: house prices; momentum; AR(2) process; rational expectations; news shocks.

Abstract:
This report presents the main results and underlying methodology of the 2021 wave of the Luxembourg Household Finance and Consumption Survey (LU-HFCS). From 2018 to 2021, there was a broad-based increase in gross and net household income, as well as in net wealth. This partly reflects the rapid increase in property prices, positive stock market developments
and “involuntary” savings during the COVID-19 pandemic. The share of indebted households remained stable, but the average amount of debt increased substantially across all household groups. There were no significant changes in income or wealth inequality since 2018.
Keywords: Household, survey, wealth, income, assets, debts, COVID-19.
JEL-Codes: D31, D14, C81, C83.

On average, consumer debt per household is twice as high in Luxembourg as in the euro area. Among lower-income households, consumer debt is even three times higher in Luxembourg. However, since incomes are also higher in Luxembourg, the ratio of consumer debt to gross income is comparable in Luxembourg and the euro area. This paper uses household survey data to compare the prevalence of consumer debt in Luxembourg and the euro area. It focuses on the two major components of consumer debt, install-ment loans and credit card debt, linking the probability of contracting these types of debt to individual household socio-economic characteristics. In the euro area, households with mortgage debt are more likely to take out con-sumer debt, highlighting the need to better understand this behavior and its potential link to financial vulnerability. Credit cards, instead, are more common in Luxembourg than in the euro area, but the share of households holding credit card debt is similar. Many euro area households that hold credit card debt also hold liquid assets, often in amounts sufficient to repay this debt. Credit constraints and differences in individual risk preferences may help to explain this otherwise puzzling behavior.
Jel Codes: G51; D14; D91; E21; G02
Keywords: household finance - consumer debt - installment loans - credit cards

Using household survey data for Luxembourg, we find that the probability of homeownership among households in Luxembourg is positively related to private wealth transfers. The size of the transfers matters. While transfers below €100,000 exert no significant influence, larger transfers increase the probability of homeownership by more than 10 percentage points. In addition, a large share of home[1]owners who received such transfers purchased their homes within a short timeframe surrounding the transfer event.

Keywords: Private wealth transfer, homeownership, household. JEL-Codes: G51, D12, D14, D31

High inflation and rising interest rates could increase financial vulnerability among Luxembourg households, who tend to carry significant debt. This paper uses micro-data on individual households from the Luxembourg Household Finance and Consumption Survey (HFCS) to identify pockets of financial vulnerability across the resident population. We calculate seven standard debt burden indicators and simulate their evolution through the end of 2023 based on BCL macroeconomic projections. According to several indicators, the share of financially vulnerable households increased from 2018 to 2021. Our simulations suggest this trend continued in 2022 and 2023 for those indicators that focus on the risk of delayed payment, although it may have reverted for other indicators that focus on the level of debt. Risk is concentrated among low-income households, but indebted households in this group only account for 11% of the number of mortgage contracts and 9% of aggregate household sector debt.

JEL-codes: E44, E47, G21, G28, G51 Keywords: Household debt; Financial vulnerability; Interest rates; Micro-simulation

Abstract. Artificial Neural Networks (ANNs) are powerful tools that can solve dynamic
programming problems arising in economics. In this context, estimating ANN
parameters involves minimizing a loss function based on the model’s stochastic functional
equations. In general, the expectations appearing in the loss function admit
no closed-form solution, so numerical approximation techniques must be used. In this
paper, I analyze a bias-corrected Monte Carlo operator (bc-MC) that approximates
expectations by Monte Carlo. I show that the bc-MC operator is a generalization of
the all-in-one expectation operator, already proposed in the literature. I propose a
method to optimally set the hyperparameters defining the bc-MC operator and illustrate
the findings numerically with well-known economic models. I also demonstrate
that the bc-MC operator can scale to high-dimensional models. With just a few minutes
of computing time, I find a global solution to an economic model with a kink in
the decision function and more than 100 dimensions.
Keywords: Dynamic programming, Artificial Neural Network, Machine Learning, Monte
Carlo.

Abstract. Long-term care (LTC) expenditures of the elderly are high in developed coun- tries and will grow further with population aging. In addition, LTC costs are heterogeneous across individuals and unknown early in life. In this paper, we add uncertainty over the arrival and magnitude of future LTC costs into a life-cycle model with endogenous aging, and we analyze how this affects the optimal behavior of agents. We show that, compared to a deterministic model, uncertainty boosts precautionary savings, lowers investment in pre- ventive care, and weakens the effectiveness of subsidies to encourage prevention. Our results therefore suggest that uncertainty should not be ignored in models that study positive or normative aspects of health investment.
JEL Codes: C60, D15, D81, I12, I18.
Keywords: health; long-term care costs; uncertainty; stochastic model.

This paper uses long-term population projections to study the evolution of dementia in Luxembourg through 2070, as well as its impact on public expenditure through healthcare and long-term care. We extend the Giordana and Pi Alperin (2022) model by adding an algorithm to identify individuals suffering from dementia. This allows us to simulate dementia prevalence among individuals aged 50 and more in several scenarios incorporating alternative hypotheses about risk factors, new treatments and comorbidities (including long-run effects of COVID-19). Public health policies reducing stroke and hypertension risk could lower dementia prevalence by 17% and public expenditure on healthcare for dementia patients by a similar amount. A new treatment extending the mild dementia phase could nearly double prevalence and possibly triple the associated healthcare costs. Finally, past exposure to COVID-19 could raise prevalence by 12% to 24% in the medium term and public expenditure on dementia healthcare by 6% to 12%. Public expenditure on long-term care for dementia patients would increase even more, generally doubling by 2070.

Keywords: Dementia, Dynamic micro-simulation, Healthcare, Health-related public expenditure, Long-term care, Luxembourg, SHARE.
JEL classification: D3, H30, I10, I12, I13, I18

Abstract This paper presents alternative distributions of foreign direct investment (FDI) stocks held by captive financial institutions (CFIs) affiliated to private equity and real estate investment funds in Luxembourg. The study focuses on CFIs with total assets larger or equal to 500 million euros. The paper breaks down FDI stocks by geographical location, sector and main economic activity. The analysis is undertaken on both the liabilities side and the assets side of CFIs’ balance sheets. The paper considers four alternative counterpart concepts: the immediate counterpart country (traditionally favoured by international statistical standards), the sponsor (or capital manager) counterpart, the target (or capital recipient) counterpart and the client (or capital provider) counterpart. Results show that the traditional distribution of FDI stocks based on the immediate counterpart country differs substantially from the alternative distributions based on the sponsor/target/client counterparts. The alternative distributions provide two important improvements. First, the traditional distribution can be misleading as blurring the initial provider of capital and the final recipient of capital, but the alternative distributions can disentangle capital providers (clients), capital managers (sponsors) and final capital recipients (targets). Second, the alternative distributions can distinguish between FDI that effectively benefits the host country and FDI that transits via the host country to be invested in third countries. This second aspect is particularly important for countries hosting a financial centre such as Luxembourg.

Keywords: Capital flows, Foreign direct investment, Captive financial institutions and money lenders, Sector S127, Investment funds, Private equity investment funds, Real estate investment funds, Financial structuring

JEL codes: C80, C81, F23, F30, G23, G32

Abstract. Not so well. We reach this conclusion by evaluating the empirical performance of a benchmark DSGE model with real estate and collateral constraints. We estimate the model from U.S. data using Bayesian methods and assess its fit along various dimensions. We find that the model is strongly rejected when tested against unrestricted Bayesian VARs and cannot replicate the persistence of real estate prices and various comovements between aggregate demand, real estate prices, and debt. Performance does not improve with alter-native definitions of real estate prices, estimation samples, or detrending approaches. Our results raise doubts about the ability of current DSGE models with real estate and collat-eral constraints to deliver credible policy insights and identify the dimensions in need of improvement.
JEL Codes: C52, E32, E44.
Keywords: real estate; housing; DSGE models; collateral constraints; model evaluation.

I explore the links between commuting costs and local employment dynamics using a spatial discontinuity introduced by a French reform in September 2015. The reform decreased the cost of public transportation in selected areas of the Paris region, but did not affect other areas. In the baseline regression framework, which only includes units that are geographically close to each other, I find that areas benefiting from the reform experienced a 0.25 percentage point decline in the unemployment rate, a 0.60 percentage point increase in the share of employed workers commuting using public transport, and a 1.4% increase in the price of residential real estate. I extend the regression framework to take into account the heterogeneity of treatment introduced by the reform, which allows me to analyze the mechanisms driving the results. I also show that a calibrated spatial search-and-matching model can rationalize the estimated treatment effects.

JEL Codes: E24, J68, R13, R23.

Keywords: Local employment, Commuting Costs, Policy, Search-and-Matching.

This paper examines the role of Luxembourg in the international banking system through the Locational Banking Statistics compiled by the Bank for International Settlements. Across European countries, Luxembourg features the largest cross-border banking positions relative to GDP. Indeed, Luxembourg is a small open economy with an international financial centre, whose banking sector consists mostly of foreign-controlled banks. The cross-border banking positions focus on loans and deposits between banks and notably intragroup positions. The geographical counterparts of cross-border banking positions in Luxembourg are mainly Western European countries (especially the euro area) and North America (notably the United States), whether for claims or liabilities. By order of importance, the main country counterparts are Germany, France, Great Britain, Switzerland, Italy, the United States, the Netherlands and Belgium. Within the international banking network, the importance of cross-border banking positions in Luxembourg resembles that of Belgium, Ireland, Japan and the Netherlands. These countries feature fewer connections than the United States, Germany and France. At the top of the network, Great Britain stands as the leading international banking centre. The structure of the international banking network evolves over time. During periods of financial stress, the density of connections stagnates or diminishes and the network becomes less resilient. This was notably the case during the global financial crisis of 2007-2008 and the European sovereign debt crisis of 2010-2012. Over time, the international banking network became more fragmented with more communities developing. This suggests a regionalisation of cross-border banking flows, as cross-border banking activity becomes more concentrated within specific groups of countries.

Keywords: Cross-border banking positions, BIS Locational Banking Statistics, Network
analysis

JEL codes: F30, E50

Empirical evidence identifies shortages of specialized labor as one of the main obstacles to technology adoption. In this paper, we explain this phenomenon by developing a model in which firms require specialized labor to produce with a new (more efficient) technology. We assume that the cost of specializing labor increases with the efficiency gains that can be attained through the new technology. This reveals two opposing effects on the endogenous share of specialized labor. On the one hand, there is a wage effect by which efficiency gains widen the wage gap between specialized and unspecialized workers, raising the share of specialized labor. On the other hand, there is a learning effect by which efficiency gains increase specialization costs, reducing the share of specialized labor. We show the learning effect will dominate when firms have sufficient market power.

Jel Codes: O33; J24; I26 Keywords: Technology adoption - education - product differentiation

Abstract. I model two ways climate tail risk could threaten the resources available for consumption in an otherwise standard cake-eating problem. I show that precautionary behaviour is optimal, no matter how low the probability of catastrophic  climate outcomes.
JEL Codes: E20, Q50.
Keywords: Climate change, Tail Risk, Tipping Point.
Summer,

Abstract. Hamilton (2018) argues that one should never use the Hodrick-Prescott (HP) filter, given its drawbacks and the existence of a better alternative. This comment shows that the main drawback Hamilton finds in the HP filter, the presence of filter -induced dynamics in the estimate of the cyclical component, is also a key feature of the alternative filter proposed by Hamilton. As with the HP filter, the Hamilton filter applied to a random walk extracts a cyclical component that is highly predictable, that can predict other variables, and whose properties reflect as much the filter as the underlying data-generating process. In addition, the Hamilton trend lags the data by construction and there is some arbitrariness in the choice of a key parameter defining the filter. Therefore, a more balanced assessment is that the HP and Hamilton filters provide different ways to look at the data, with neither being clearly superior from a practical perspective.

JEL Codes: B41, C22, E32.

Keywords: HP filter; Hamilton filter; business cycles; detrending; filtering.

In 2019, Luxembourg introduced borrower-based instruments in the macro-prudential toolkit to constrain credit to households who exceed a certain limit on their loan-to-value ratio, on their (mortgage) debt-to-income ratio or on their debt service-to-income ratio. This paper analyses the impact of setting these limits at different levels, using household-level data from Luxembourg. We calculate these debt burden ratios for individual households who recently purchased their main residence using data from the Household Finance and Consumption Survey conducted in 2010, 2014 and 2018. On January 1, 2021 authorities imposed a legally binding limit on the loan-to-value (LTV) ratio for new mortgages. This may be 80%, 90% or 100% depending on the category of borrower. Had the least restrictive LTV limit envisaged by the law (100%) been applied in 2018, credit would have been rationed to 24% of households with recent mortgages on their main residence. This limit would have required a 7% reduction in the overall debt of this group of households. Had the most restrictive LTV limit envisaged by the law (75%) been applied in 2018, credit would have been rationed to 64% of households with recent mortgages on their main residence, requiring an 18% reduction in overall debt in this group. More generally, we evaluate how well borrower-based instruments can target those households that are financially vulnerable (according to conventional measures from the literature). By simulating an adverse scenario, we find that combining several ratios one could better target households that were not financially vulnerable in the benign conditions of 2018 but would become vulnerable after a shock to income. However, any borrower-based instrument inevitably generates some classification errors (either granting credit to households that are financially vulnerable or constraining credit to households that are not financially vulnerable). Using different assumptions on policymaker preferences, we apply the signals approach to derive limits that are “optimal” in the sense of minimising classification errors.
JEL-codes: D10, D14, G21, G28
Keywords: Household debt; Financial vulnerability; Macro-prudential policy

Reliable macroeconomic housing and wealth statistics as well as counterfactual analyses across housing tenure status require hypothetical sales and rent prices for properties off the market reflecting current market conditions and representing the entire housing stock. We replace subjective values reported by participants in the Luxembourg Household Finance and Consumption Survey by objectified values imputed via hedonic models estimated on observable market data. We find that the participants’ tendency to over- and under-report values is strongly correlated with tenure length, tenure type, dwelling type, household income and wealth. We find shifts in the wealth distribution, detect large regional variation in price-to-rent, price-to-income and rent-to-income ratios as well as significant affordability concerns: only 18% of all renting households could theoretically afford to purchase the dwelling they currently rent at market conditions. These renters are usually younger, at the top of the wealth and income distribution, and reside outside Luxembourg City.

Keywords: Macroeconomic Statistics; Subjective Assessments; Surveys; Measurement Errors; Housing and
Rent Markets; Housing Wealth; Affordability
JEL codes: E58; G51; R21; R31.

Abstract. We develop a monocentric urban search-and-matching model in which workers can choose to commute or to migrate within the region. The equilibrium endogenously allocates the population into three categories: migrants (relocate from their hometown to the city), commuters (traveling to work in the city) and home stayers (remaining in their hometown). We prove that the market equilibrium is usually not optimal: a composition externality may generate under- or over-migration with respect to the central planner’s solution, which in all cases results in under-investment in job vacancies and therefore production. We calibrate the model to the Greater Paris area to reproduce several gradients observed in the data, suggesting over-migration. We show how policy interventions can help to reduce inefficiencies.

JEL Codes: E24, J68, R13, R23.

Keywords: Migration, Commuting, Urban search-and-matching, Efficiency, Policy.

Abstract: This paper simulates long-term trends in Luxembourg’s public expenditure on healthcare and on long-term care. We combine population projections with micro-simulations of individuals’health status that account for their demographic, socio-economic characteristics and their childhood circumstances. Model equations estimated on data from the SHARE survey and from several branches of Social Security provide a rich framework to study policy-relevant applications. We simulate publicexpenditure on healthcare and long-term care under different scenarios to evaluate the separate contributions of population ageing, costs of producing health-related services, and the distribution of health status across age cohorts. Results suggest that rising per capita expenditure on healthcare will mostly result from production costs, while rising expenditure on long-term care will mostly reflect population ageing.

Keywords: Ageing, Dynamic micro-simulation, Healthcare, Health-related public expenditure, Health

status, Long-term care, Luxembourg, SHARE.

JEL Classification: D3, H30, I10, I12.

The paper draws a typology of captive financial institutions and money lenders (CFIs, sector S127) in Luxembourg from a new database. The latter retrieves information from three sources: the EuroGroups Register managed by Eurostat, the Statistical Business Register managed by the STATEC (National Institute for Statistics and Economic Studies) and the Central Balance Sheet Register managed by the STATEC. The new database enhances the data coverage of CFIs in Luxembourg. Indeed, it includes not only CFIs with total assets larger or equal to EUR 500 million as in the BCL reporting (BCL (2014)), but also CFIs whose total assets are lower than EUR 500 million. The period of analysis spans 2011 to 2019. Results show that CFIs present different characteristics depending on their balance sheet size. On the one hand, CFIs with total assets larger than EUR 100 million mainly regroup holding companies, intragroup lending corporations, mixed structures and conduits. On the other hand, CFIs with total assets lower than EUR 100 million feature mostly mixed structures. Overall, while holding corporations own the majority of total assets, the largest number of companies consists of mixed structures.

Keywords: Captive financial institutions and money lenders, Sector S127, Typology, EuroGroups Register, Statistical Business Register, Central Balance Sheet Register, Big Data

This paper analyses the effects of the COVID-19 pandemic shock on small open economies in a monetary union with an application to the euro area. Accounting for a high degree of openness and a strong dependence on intra and extra union trade, we focus on the size and the direction of international spillovers – both from the shock itself and from the ensuing fiscal response. To do so, we use a unified modelling framework: The Euro Area and the Global Economy (EAGLE) model. Furthermore, within this general framework, we assess the extent to which specific modelling features shape the dynamic responses to the COVID- 19 pandemic. The main messages are as follows. First,  fiscal spillovers from the rest of  the monetary union do matter. Second, the effective lower bound amplifies the size of the spillovers. Third, the design of wage negotiations leads to wage subsidies having negative international fiscal policy spillovers. Fourth, import content of government spending interacts with the effective lower bound, strongly affecting the size and sign of spillovers. Fifth, when households have finite lifetimes, the responses of output and inflation are amplified compared to the case with infinitely lived households. Finally, a next generation EU instrument is more effective when financed using a tax on consumption.

JEL Classification: C53, E32, E52, F45.
Keywords: DSGE Modelling, International Spillovers, Monetary Union, Euro Area, COVID-19

Abstract: The introduction of a general-purpose central bank digital currency (CBDC) carries the risk of bank disintermediation, potentially jeopardizing financial stability and monetary policy transmission through the bank lending channel. By adapting the theoretical framework of Dutkowsky and VanHoose (2018b, 2020) to the euro area, this study clarifies the conditions under which a digital euro could be introduced on a large scale without leading to bank disintermediation or a credit crunch. First, the central bank would need to set up proper mechanisms to manage the volume and the user cost of CBDC in circulation. Second, since some bank deposits will be converted into CBDC, the central bank should continue to facilitate access to its long-term lending facilities in order to provide banks with an alternative funding source at an equivalent cost. Depending on its design, a digital euro could improve bank profitability by absorbing large amounts of idle (and expensive) excess reserves without penalizing lending. A digital euro could also improve banks’ competitive position relative to non-bank lenders and encourage bank digitalization.

JEL Classification: E41, E42, E51, E52, E58, G21
Keywords: Central bank digital currency, cash, central bank, monetary policy, excess reserves, reserve
requirements, universal central bank reserves, bank deposits, bank profitability, bank credit, inside money, collateral

This report presents the methodology and main descriptive results from the third wave of the Cross-border Household Finance and Consumption Survey (XB-HFCS) conducted in 2018. This is a household survey of employees in Luxembourg who live abroad and regularly commute across the border. We analyse the composition and level of household assets and liabilities, net wealth and income, and compare them to those of similar households (including at least one employee) whether resident in Luxembourg or in one of the bordering countries. Compared to households employed in their country of residence, cross-border commuters reported higher median gross income, homeownership rates and wealth. Around 10% of cross-border commuters lived in Luxembourg before they moved to Belgium, France or Germany (usually their country of birth).

Keywords: Cross-border commuters, households, survey, assets, liabilities, wealth, income.

JEL-Codes: G51, D31, D14, C81, C83, J61

The uncertainty that surrounds public perceptions of central bank policy matters for business cycle dynamics. An increasing number of empirical studies document the contractionary effects of increases in monetary policy uncertainty in developed economies.
I offer a theoretical explanation for these contractionary effects. To this end, I present a New Keynesian model in which private agents have limited information about the central bank’s reac-tion function. The latter alternates between periods of active inflation stabilization and periods of passive inflation stabilization. Each quarter, private agents estimate the parameters of the current monetary regime by combining what they see happening in the economy with their own past beliefs.
This belief structure accounts for the contractionary effects of increases in monetary policy uncertainty. Spikes in uncertainty about the current monetary regime obscure the likely path of future nominal interest rates, and hence, of the marginal product of capital. This unpredictability encourages private agents to defer investment decisions until additional information becomes available. The fall in investment then hinders economic activity through the standard channels: a decline in aggregate demand, and the prospect of a protracted period of unusually low capital stock.
In addition, I explore how uncertainty about the current monetary policy regime affects the transmission of monetary policy shocks. By softening the link between fundamentals and equi-librium prices and allocations, learning renders the economy less responsive to monetary policy.
Changes in the monetary policy regime in the model capture a constrained discretion strategy [Bernanke and Mishkin (1997)], followed by the Federal Reserve and other central banks. This allows the central bank some flexibility to de-emphasize inflation stabilization to pursue alter-native short-term goals, limiting cyclical swings in resource utilization, while retaining a strong commitment to keeping inflation low and stable.

Abstract. The joint behavior of Total Factor Productivity (TFP) and the Relative Price of Investment (RPI) in the data lead several authors to conclude that neutral technology shocks are positively correlated with investment-specific technology shocks, challenging the specification of standard macroeconomic models. This paper rejects the correlated-shocks hypothesis using both parametric and non-parametric methods and controlling for structural breaks. The data suggests moderately negative long-run covariation between the RPI and TFP constructed from chain-linked output, but the RPI is orthogonal to TFP in consump-tion units. These results are consistent with a simple two-sector model in which neutral technology shocks and investment-specific technology shocks are uncorrelated, while mod-els with correlated shocks cannot account for the second result. I conclude that it is not necessary to adapt macro models to allow for correlated technology processes
JEL Codes: E30, E32, O41.
Keywords: total factor productivity, relative price of investment, neutral technology, invest-ment-specific technology, long-run covariability, structural VARs.

We study how globalization affects the response of mutual fund flows to past performance. For that purpose, we use a novel dataset on bond funds from the internationalized Luxembourg fund industry. We find that flows into global funds, i.e. funds issuing shares in several currencies, are more sensitive to past performance than flows to domestic funds, i.e. funds distributing shares in mainly one currency. Moreover, global funds exhibit a higher flow sensitivity to low and high performance, while flows to domestic funds are more reactive to medium performance. These results are robust to using alternative measures of globalization to define domestic and global funds, like the geographical diversification in the distribution of shares and the geographical and currency diversification in the asset portfolio. Thus, the globalization dimension of mutual funds, neglected by related studies, raises the sensitivity of flows to past performance and needs to be taken into account by supervisory and regulatory authorites.

KEYWORDS: Mutual Funds, Multi-currency issuance, Globalization, Flow-performance relationship
JEL CODES: F30, G11, G23

The paper provides insights into captive financial institutions and money lenders (sector S127) in Luxembourg. To this end, the paper relies on two metrics: the total assets and the total number of S127 entities. The analysis breaks down the evolution of both metrics across various dimensions: by main economic activity performed by the groups owning S127 entities, by nationality of the groups, by geographical counterpart of the balance sheet of S127 entities, by balance sheet item and by type of S127 firm. The period covers Dec. 2014 - Dec. 2019. Given data availability, the analysis relies on a sub-sample of the whole population of S127 firms. This sub-sample features S127 firms whose total assets are at least equal to EUR 500 million. As of Dec. 2018, this sub-sample represents about 5% of the total number of S127 firms in Luxembourg, and about 85% of the total assets held by S127 firms in Luxembourg. Results show that groups resorting to S127 entities perform various economic activities. Groups undertaking non- financial activities account for 80% of the total assets held by S127 entities and 70% of the total number of S127 entities. The remaining share relates to groups involved in financial activities. Across economic activities, the finance and insurance industry - and notably investment funds - holds the most important share of assets and number of entities in the sample of S127 firms. Groups headquartered in the United States own the major share of S127 entities, whether in terms of total assets or number of S127 entities. Luxembourg represents the most important balance sheet counterpart of S127 entities, suggesting that groups own several captive financial institutions in Luxembourg. The predominant financing means in the aggregated balance sheet of S127 entities mainly relate to equity and debt both as direct investment. The most important type of S127 entities are holding entities, followed by intragroup lending corporations and mixed structures. The dynamic analysis shows that total assets held by S127 entities increased over the period Dec. 2014 - June 2017 and decreased from July 2017 to Dec. 2019. The total number of S127 entities rose over the period Dec. 2014 - Nov. 2018 and declined between Dec. 2018 and Dec. 2019. The main contributors underlying these dynamics differ in terms of magnitude and sign over time, depending on the main economic activity performed by the groups owning S127 entities, the nationality of the groups, the balance sheet counterpart of S127 entities, the balance sheet item and the type of S127 firm. The main motive driving the expansion phase of the number of S127 entities before Dec. 2018 relates to entities created in the past and whose total assets increased above the EUR 500 million reporting threshold. The main motives driving the decrease in the number of S127 entities after Dec. 2018 pertain to S127 entities whose total assets fell below the EUR 500 million reporting threshold but that are still operating, and to liquidations of S127 entities.

Keywords: Captive financial institutions and money lenders, Sector S127, Multinational Enterprises (MNEs), Ownership chains, Organizational structure

JEL codes: C80, C81, L22

International capital mobility intensifies tax competition between jurisdictions. However, many firms only operate domestically and are internationally immobile. This paper aims to analyze the effect of tax competition on the profitability of local (immobile) firms, especially when tax and non-tax instruments, including infrastructure provision, are involved. We show that tax competition decreases investment and profit of local firms when internationally mobile firms do not benefit sufficiently from local infrastructure.

Using a dedicated set of questions in the 2014 Luxembourg Household Finance and Consumption Survey (LU-HFCS), we show that a substantial share of households contributes their own labour to the acquisition of their main residence. These contributions help households faced with credit constraints, since they reduce the need for external financing. We develop a simple theoretical model and show that own labour contributions decrease with the level of financial resources available, while they increase with the mortgage interest rate. These theoretical results are supported by empirical analysis, which also shows that own labour contributions vary by household characteristics (age, gender, profession) and by type of dwelling (house, apartment). 

JEL Codes:  D14, E43, G21, R21

Keywords:  Borrowing constraints, down payment, own labour, sweat equity, homeownership, household, survey.

Abstract. This paper outlines a new estimated dynamic stochastic general equilibrium (DSGE) model of the Luxembourg economy named LED, for Luxembourg Estimated DSGE. The paper provides a thorough discussion of the model structure, explains how LED is solved and estimated, and shows how it can be used to study important properties of the Luxembourg economy. The empirical results are encouraging: parameter estimates take reasonable values, the model fits the data well, and its implications regarding the determinants of economic growth and cyclical fluctuations in Luxembourg are plausible.

JEL Codes:  C11, C32, E32, E37.

Keywords: DSGE models, open-economy macroeconomics, Bayesian inference, policy anal- ysis, Luxembourg.

The paper presents a typology of captive financial institutions and money lenders (sector S127) in

Luxembourg. Given data availability, the analysis relies on a sub-sample of the whole population of S127 firms. This sub-sample features S127 firms whose total assets are at least equal to EUR 500 million. As of Q4 2018, this sub-sample represents about 5% of the total number of S127 firms in Luxembourg, and about 85% of the total assets held by S127 firms in Luxembourg. The period of analysis spans Q4 2014 to Q4 2019. In terms of number and on average over the period Q4 2014 – Q4 2019, the sample of S127 corporations regroups holding corporations (42%), intragroup lending companies (25%), mixed structures (19%), conduits (7%) and loan origination companies (4%). These corporations represent about 98% of the total number of S127 companies whose total assets of at least EUR 500 million. The remaining types that complete the sample of S127 entities consist of captive factoring and invoicing corporations, companies with predominant non-financial assets, extra-group loan origination firms, wealth-holding entities and captive financial leasing corporations. In addition, on average over the period Q4 2014 – Q4 2019, holding corporations own the largest share of total assets (55%) followed by intragroup lending companies (22%), mixed structures (14%), conduits (6%) and loan origination companies (2%). These corporations account for about 99% of the total assets held by S127 companies whose total assets are at least equal to EUR 500 million. The relative importance of holding corporations, intragroup lending companies, mixed structures, conduits and loan origination companies suggests that Luxembourg plays the role of a global financial centre for MNEs. The latter benefit from Luxembourg as a financial platform to manage their business activities and structure their corporate investments.

Keywords: Captive financial institutions and money lenders, Sector S127, Typology

JEL codes: C80, C81, L22

The ratio of gross debt-to-GDP for non-financial corporations (NFCs) is often used to assess corporate indebtedness. Across European Union (EU) countries, Luxembourg holds the largest ratio. A natural question that arises is whether there should be any concern in terms of excessive debt build-up for NFCs in Luxembourg. Against this background, the paper analyses the underlying characteristics of NFC debt in Luxembourg. It undertakes a macroeconomic and a firm-level analysis. It explores the debt components, their counterparts and the type of NFCs that contribute to the debt. The paper comes up with alternative and complementary indicators aimed to provide a finer assessment of NFC debt in Luxembourg.
The macroeconomic analysis shows that the main components of NFC debt in Luxembourg are loans granted whether from resident NFCs or from resident/non-resident captive financial institutions and money lenders. On the asset side, debt mainly finances corporate investments in the form of unlisted shares (to resident NFCs and resident/non-resident captive financial institutions and money lenders) or loans (to resident/non-resident NFCs or to resident captive financial institutions and money lenders).
The firm-level analysis shows that debt issued by foreign-controlled NFCs predominates over debt issued by national NFCs. In other words, debt issued by foreign-controlled NFCs represents the major part of NFC debt at the macroeconomic level. The largest component of debt for foreigncontrolled NFCs are intra-group loans. Another important observation is that between types of NFCs, debt is not equally distributed across firms. This is particularly true for foreign-controlled NFCs. This suggests that some companies contribute more than others to the large ratio of debt-to-GDP.
Altogether, the macroeconomic and firm-level analyses provide evidence that the large ratio of NFC debt-to-GDP for Luxembourg compared to other EU jurisdictions stems from a structural feature of Luxembourg that pertains to its role as a global financial center. Indeed, the country hosts a large number of NFCs and notably foreign-controlled NFCs (including large multinational enterprises) that benefit from Luxembourg as a financial platform to manage their business activities and structure their corporate investments.
While the ratio of debt-to-GDP places Luxembourg NFCs as the largest holders of debt across EU countries, alternative indicators suggest the opposite result. This is notably the case of the ratio of debt-to-financial assets at the macroeconomic level as Luxembourg NFCs hold the largest stock of financial assets across EU countries. A similar observation holds for the ratio of debt-to-total assets whether for foreign-controlled NFCs, national NFCs or when considering the sector of NFCs as a whole.

The rate of homeownership in Luxembourg is close to the OECD average. However, strong house price increases, mainly driven by population growth and limited housing supply, reduce housing affordability, in particular for the young, and contribute to the net wealth gap between homeowners and renters. As in many OECD countries, housing is the main asset of the middle class. However, at the top of the wealth distribution housing is less prominent and accounts for a smaller share of wealth than in most OECD countries. Mortgage market participation in Luxembourg is higher than in neighbouring countries and households in the middle income quintile are almost as likely to have a mortgage as those in the top income quintile. Among non-resident commuters (who cross the border every day to work in Luxembourg), homeownership is higher than the average for the country in which they live, mainly reflecting their higher income. Still, commuters often identify high real estate prices as the reason for not moving to Luxembourg. Among Luxembourg residents, a third are renters, often citing high real estate prices and insufficient own funds as obstacles to homeownership. Even controlling for other household characteristics, there is a substantial gap in net wealth between renters and homeowners. The data also indicates that median net wealth among Luxembourg residents is significantly higher than among cross-border commuters. For Luxembourg employed residents and cross-border workers from different countries, the empirical analysis confirms that higher education and income play an important role in explaining wealth differences between households.

Keywords: Household, survey, wealth, income, assets, debt, cross-border commuters.

JEL-Codes: D31, D14, C81, C83.

TFP measures constructed from chain-aggregated output, such as those published by the Bureau of Labor Statistics or Fernald (2014), confound contributions from neutral and sector-specific technology. Therefore, they should not be used to infer the path of neutral technology in presence of investment-specific technical change. Two theory-consistent, utilization-adjusted measures of neutral technology at the quarterly frequency are proposed for the US business sector. Both indicate that neutral technology progress declined dramatically after the mid-1970s. In particular, its contribution to US growth fell from more than 85% before 1973 to less than 25% afterward. The associated welfare loss is enormous: if neutral technology had continued on its pre-1970s trend, 2017 US output would have been 70% higher.

JEL Codes: E22, E23, E32, O41, O47.

Keywords: total factor productivity, neutral technology, investment-specific technology,
sources of growth.

This report presents the main results and underlying methodology of the third wave of the Luxembourg Household Finance and Consumption Survey (LU-HFCS) and compares them to results obtained in the first and second wave in 2010 and 2014. This survey is conducted among private households resident in Luxembourg and is part of the Eurosystem Household Finance and Consumption Survey, which provides detailed individual and household data on assets, liabilities, income and consumption. This individual-level information on households provides a view on the distribution of assets and liabilities across the population that complements the aggregate data on the household sector in the financial accounts.

Keywords: Household, survey, wealth, income, assets, debts.

JEL-Codes: D31, D14, C81, C83.

This paper assesses ‘aggregate vulnerability’, a measure of systemic risk, in the investment fund sector of Luxembourg by implementing a macroprudential stress testing model. While based on the proposal by Fricke and Fricke (2017), this paper focuses on the calibration of key parameters such as the flow-performance sensitivity and price impacts that are included in the model to capture the so-called ‘second-round effects’ of an initial adverse shock to funds’ returns. According to the empirical results, limited degrees of vulnerability were found for the main fund categories such as equity funds, bond funds and mixed funds. This implies that the investment fund sector in Luxembourg does not raise any particular concern for financial stability as of November 2019. However, since the stress test was performed against a background of increased risk of a reversal in global risk premia, continued monitoring of the sector is warranted.

Keywords: investment funds; macroprudential stress test; flow-performance sensitivity; price impact; fire sales; systemic risk

JEL Classification: G11, G12, G23

This document presents the structure of PENELOPE, a macro-accounting tool aimed at examining the long-run sustainability of the Luxembourg pension system. PENELOPE complements othermodels studying the Luxembourg pension systemby proposing a disaggregated pension analysis in which demographic changes affect macroeconomic variables. The results of PENELOPE’s reference scenario are compared with those of other studies, while additional simulations focus on (i) the effects of different population projections and assumptions on cross-border worker inflows, (ii) the implications of the 2012 pension reform and of alternative proposals, as well as (iii) the evolution of the pension reserve under different scenarios.

Keywords: Pension expenditure projections, demographic trends, labor inflows, pension reform

JEL-Codes: H55, H68, J11

Health subsidies involve public budgetary costs. However, they generate a positive externality by encouraging participation in health-improving initiatives, which help reduce future health care costs. We build an overlapping generations model with a government subsidizing investment in health by the young generation and paying the health care costs of the old generation. We find that the welfare-maximizing subsidy rate depends positively on health externality and the size of health care costs, and negatively on the discount factor. The subsidy rate should therefore be high when prevention more effective at cost saving and when the population is myopic about the future. Moreover, the welfaremaximizing subsidy rate is lower than the health-maximizing rate but higher than the capital-maximizing rate.

Keywords: Overlapping generations model, health subsidy, welfare.

JEL-Code: H23, I18, O41.

∗

How long does it take for exchange rate changes to pass through into in flation? Does it make a difference whether the exchange rate depreciates or appreciates? Do relatively large exchange rate changes entail more exchange rate pass-through? In this paper, we examine possible non-linearities in the transmission of exchange rate movements to import and consumer prices in all 19 euro area countries as well as the euro area as a whole from 1997 to 2019Q1. We extend a standard single-equation linear framework with additional interaction terms to account for possible non-linearities and apply local projections to obtain state-dependent impulse response functions. We fnd that (i) euro area consumer and import
prices respond significantly to exchange rate movements after one year, responding more when the exchange rate change is relatively large; and (ii) euro appreciations and depreciations afect the level of euro area exchange rate pass-through in a symmetric fashion; (iii) for euro area countries results difer for import and consumer prices and across countries.

This paper studies how peers’ financial behaviour affects individuals’ own investment choices. To identify the peer effect, we exploit the unique composition of the Luxembourg population and use the differences in stock market participation across various immigrant groups to study how they affect stock market participation of natives. We solve the reflection problem by instrumenting immigrants’ stock market participation with lagged participation rates in their countries of birth. We separate the peer effect from the contextual and correlated effects by controlling for neighbourhood and individual characteristics. We find that stock market participation of immigrant peers has sizeable effects on that of natives. We also provide evidence that social learning is one of the channels through which the peer effect is transmitted. However, social learning alone does not account for the entire effect and we conclude that social utility might also play an important role in peer effects transmission.

JEL: G5, D14, D83, G11, I22

Keywords: peer effects, stock market participation, social utility, social learning

This study examines the revision histories of national accounts data in Luxembourg. I analyse first releases and revisions in the quarterly national accounts (QNA) published by the National Institute of Statistics (STATEC). Reliability is evaluated by measuring revision size, variability as well as the frequency in sign changes and acceleration/deceleration switches. In addition, the predictability of revisions is assessed by applying regression analysis. Overall, the results point to high uncertainty surrounding early QNA estimates, also in international comparison. I find that revisions to GDP and its components are substantial. While there is no clear evidence of a bias in year-on-year real GDP growth, this does not
hold for some GDP components.

Abstract. We augment the original LU-EAGLE model with disaggregated public expenditure, allowing for (i) a distinction between public consumption and investment expenditures, (ii) complementarity between public and private consumption, (iii) a productive role for public capital, and (iv) separate private and public employment. This extended model embeds a wide range of transmission channels from public expenditures and allows for a detailed analysis of the general-equilibrium effects of public demand in Luxembourg. Model simulations suggest that a rise in public employment induces the strongest GDP response in the short run, while a rise in public investment has the largest effects in the long run. The results also indicate that crowding-out effects through changes in net exports are essential in determining fiscal multipliers for small open economies such as Luxembourg.
JEL Codes: C54, E17, E32, E37, E62, F47.
Keywords: DSGE models, open economy models, fiscal policy, Luxembourg.

We augment a simple Real Business Cycle model with financial intermediaries that may default on their liabilities and a financial friction generating social costs of default. We provide a closed-form solution for the general equilibrium of the economy under specific assumptions, allowing for analytic results and straightforward simulations. Endogenous default generates asymmetric business cycles and our model replicates both the negative skew of GDP and the positive skew of credit spreads found in US data. Stronger financial frictions cause a rise in asymmetry and amplify the welfare costs of default. A Pigouvian tax on financial intermediation mitigates most of these negative effects at the cost of a steady-state distortion.

JEL Codes: E32, E44, G21

Keywords: Real Business Cycle model, default, financial frictions, asymmetry, skewness.

This paper studies the cyclical properties of bank loans to non-financial corporations, bank loans to households, bank loans to the non-financial private sector and house prices in Luxembourg, by applying two methodologies to decompose and characterize the cycles. First, we use an unobserved components model (UCM) to extract classical cycles. We find evidence of medium-term cycles in the loan series over the sample 1980 Q1 to 2019 Q1. The nonfinancial corporation credit cycle is the most volatile and of larger duration compared to the other credit cycles. The length of the house price cycle in Luxembourg is estimated at 13.8 years. A dynamic synchronicity measure between households’ credit cycle and the house price cycle reveals that these cycles are sometimes synchronous. However, the early warning properties of the univariate unobserved components model are limited despite its good pseudo real-time estimates. A wavelet analysis complements the findings of the univariate UCM by providing growth cycles of financial variables. We show that these growth cycles could be useful as complementary information in financial cycle analysis. A coherence wavelet analysis is also conducted. The main findings show that the credit growth cycles of non-financial corporations and of the non-financial private sector are highly coherent with respect to households’ credit growth cycle. The household credit cycle often precedes house price growth. From a macroprudential policy perspective, these results support the use of complementary methodologies for assessing the financial cycle and confirm earlier studies on the limited role of the household credit cycle in the evolution of the house price cycle.

Keywords: credit cycle, wavelet analysis, unobserved components model, house prices

JEL codes: C30, E32, E51

Abstract

This paper models how internet platforms decide whether to introduce virtual currencies. Since platforms operate two-sided markets, virtual currencies may attract users who buy goods/services as well as external firms who accept virtual currency as payment. We find that platform incentives to introduce virtual currencies depend on the distribution of wages across the population of users as well as the distribution of preferences for online activities ("digital" preferences). We use Luxembourg data from the EU Survey on Information and Communication Technologies to test model predictions on user time allocation. In particular, we identify various individual socio-economic characteristics linked to time spent on social networks. Then, we use the user net income distribution (conditional on digital preferences) to evaluate conditions determining the platform’s choice of virtual currency design.

Keywords: private virtual currencies, social networks, retail payments.

JEL classification: E42, E5, G23, L5, L82, L86.

Abstract

This paper finds that employment protection legislation (EPL) had a significant impact on employment adjustment in Europe over 2001-2013, once we account for firm-size related exemptions to EPL. We construct a novel coverage-adjusted EPL indicator and find that EPL hinders employment growth at the firm level and increases the share of firms that remain in the same size class. This suggests that stricter EPL restrains job creation because firms fear the costs of shedding jobs during downturns.

We do not find evidence that EPL has positive effects on employment by limiting job losses after adverse shocks. In addition to standard controls for the share of credit-constrained firms and the position in the business cycle, we also control for sizerelated corporate tax exemptions and find that these also significantly constrain job creation among incumbent firms.

Keywords: employment protection; firm growth; job reallocation.

JEL Codes: J08, D22.

Abstract: We analyse several motivations for the introduction of a widely accessible central bank digital currency (CBDC). If a central bank decided to offer a CBDC, its design would have to consider different areas of central bank activity, taking into account multiple policy principles, objectives and constraints. In addition, the introduction of a CBDC on a large scale may have a non-trivial impact on the architecture of the financial system. From this perspective, some common arguments in favor of CBDC may seem simplistic and the field of feasible options may be narrower than often believed.

We reconsider Tobin’s idea to establish a system of universal access to central bank reserves, and clarify its feasibility and advantages as an account-based CBDC.

JEL Classification: E41, E42, E43, E51, E52, E58

Keywords: Central bank digital currency, universal central bank reserves, deposited currency accounts, cash, central bank, central bank policies, monetary policy, financial stability, payment systems, deposit insurance, bank deposits, inside money, collateral, virtual currencies

Abstract

This study investigates the optimal macroprudential policies for Luxembourg using an estimated closed-economy DSGE model. The model features a monopolistically competitive banking sector, a collateral constraint and an explicit differentiation between the flow and the stock of household mortgage debt. Based on a welfare-oriented approach and in a context of easy monetary policy environment, we first find that the non-joint optimal loan-to-value (LTV) and risk weighted capital requirement (RW) ratios for Luxembourg seem to be 90% and 30%, respectively, while the joint optimal ratios are found to be 100% and 10% respectively. Our results from the combination of instruments suggest that the policy scenario that provides better stabilization effects on mortgage credits isn’t necessarily the one that is welfare improving. In other words, we find a complementarity between LTV and RW in terms of welfare, while their optimal combination diminishes the stabilization effects on mortgage debt and house prices. However, the time-varying and endogenous rules for LTV and RW improve the social welfare and better stabilizes mortgage loans and house prices compared to their static exogenous ratios. We further find that the optimal interactions between LTV and RW ratios in our modelling framework exhibit a convex shape. It should be recalled that the results are conditional on the model’s specific assumptions.

JEL-Classification: E32, E44, R38.

Keywords: LTV, Risk weights, optimal macroprudential policy, combination of macroprudential instruments.

Abstract
We use firm-level survey data from 25 EU countries to analyse how firms adjust their labour costs (employment, wages and hours) in response to shocks. We develop a theoretical model to understand how firms choose between different ways to adjust their labour costs. The basic intuition is that firms choose the cheapest way to adjust labour costs. Our empirical findings are in line with the theoretical model and show that the pattern of adjustment is not much affected by the type of the shock (demand shock, access-to-finance shock, ‘availability of supplies’ shock), but differs according to the direction of the shock (positive or negative), its size and persistence. In 2010-13, firms responding to negative shocks were most likely to reduce employment, then hourly wages and then hours worked, regardless of the source of the shock. Results for the 2008-09 period indicate that the ranking might change during deep recession as the likelihood of wage cuts increases. In response to positive shocks in 2010-13, firms were more likely to increase wages, followed by increases in employment and then hours worked suggesting an asymmetric reaction to positive and negative shocks. Finally, we show that strict employment protection legislation and high centralisation or coordination of wage bargaining make it less likely that firms reduce wages when facing negative shocks.

Keywords: Shocks, firms, labour cost adjustment, wages, employment, hours, survey.
JEL classification: D21, D22, D24

Cette étude vise à évaluer les risques portant sur la soutenabilité de la dette publique au Luxembourg. Elle repose sur l’utilisation d’un outil DSA (« Debt Sustainability Analysis ») développé par l’Eurosystème, dont la méthodologie est adaptée et appliquée au cas du Luxembourg. Les résultats obtenus suggèrent que, en l’absence de chocs négatifs et en considérant que le Luxembourg respecte les critères du Pacte de Stabilité et de Croissance, les risques pesant sur la soutenabilité de la dette semblent être faibles dans l’horizon considéré. Néanmoins, bien que le niveau de dette publique soit relativement bas, les résultats permettent d’identifier certains risques susceptibles d’exercer une pression à la hausse sur la trajectoire de la dette et ainsi menacer sa soutenabilité. 

Mots-clés : Dette publique, soutenabilité de la dette, DSA, gouvernance européenne.

Classification JEL : E62, H62, H68. 

We use unobserved components time series models to extract real and financial cycles for Luxembourg over the period 1980Q1-2018Q2. We find that financial cycles are longer and have larger amplitude compared to standard business cycles. Furthermore, financial cycles are highly correlated with cycles in GDP. We compare our results to other approaches to measure financial cycles and show how unobserved components models can serve to evaluate uncertainty and to monitor cyclical developments in real time. Overall, our estimates indicate that in mid 2018 both real and financial cycles in Luxembourg were close to zero, with financial conditions near their long-run trend.

JEL Codes: C22, C32, E30, E50, G01.

Keywords: financial cycles, unobserved component time series models, Luxembourg.

We argue that shocks to credit supply by shadow and retail banks were key to understand the behavior of the US economy during the Great Recession and the Slow  Recovery. We base this result on an estimated DSGE model featuring a rich representation of credit flows. Our model selects the two banking shocks as the most important drivers of the crisis because they account simultaneously for the fall in real activity, the decline in credit intermediation and the rise in lending-borrowing spreads. On the other hand, in contrast with the existing literature, our results assign only a moderate role to productivity and investment efficiency shocks.

KEYWORDS: Shadow Banking, Great Recession, Slow Recovery, estimated DSGE models.
JEL CODES: C32, E32.

Abstract

Using firm-level data from a large-scale European survey among 20 countries, we analyse the determinants of firms using short-time work (STW). We show that firms are more likely to use STW in case of negative demand shocks. We show that STW schemes are more likely to be used by firms with high degrees of firm-specific human capital, high firing costs, and operating in countries with stringent employment protection legislation and a high degree of downward nominal wage rigidity. STW use is higher in countries with formalised schemes and in countries where these schemes were extended in response to the recent crisis. On the wider economic impact of STW, we show that firms using the schemes are significantly less likely to lay off permanent workers in response to a negative shock, with no impact for temporary workers. Relating our STW take-up measure in the micro data to aggregate data on employment and output trends, we show that sectors with a high STW take-up exhibit significantly less cyclical variation in employment.

Key words: Firms, survey, crisis, short-time work, wages, recession.

JEL classification: C25, E24, J63, J68.

Abstract

We solve a version of the analytical Real Business Cycle (RBC) model with a predetermined rate of return on household saving. The solution differs from that of the benchmark RBC model along two dimensions: (i) Policy functions depend on the variance of the technology shock. (ii) There is a suboptimal pattern of excess saving. We discuss the economic intuition underlying these properties. We also demonstrate that unconditional welfare can be higher in the suboptimal model with predetermined interest rates, providing a clear illustration of the pitfall with unconditional welfare comparisons.

JEL Codes: E13, E21, E32, E43.

Keywords: RBC model, predetermined interest rates, over-saving, conditional and unconditional welfare.

Abstract. We describe LU-EAGLE, a DSGE model developed at the Banque centrale du Luxembourg. LU-EAGLE borrows its general structure from the Euro Area and GLobal Economy (EAGLE) model developed by the European System of Central Banks and also embeds specific features to capture some important characteristics of Luxembourg's economy. In particular, the model reproduces the high levels of exports and imports relative to GDP, as well as the significant share of cross-border workers in Luxembourg's labor market. We calibrate LU-EAGLE and discuss simulation results describing the effects of a set of standard shocks, originating both in Luxembourg and abroad. The model suggests that international spillovers make Luxembourg more responsive to monetary policy shocks and less responsive to fiscal policy shocks. Moreover, it highlights how fluctuations in foreign demand have a significant impact on domestic developments.

JEL Codes: C54, E17, E32, E37, E62, F47.

Keywords: DSGE models, open economy models, policy analysis, Luxembourg.

This paper uses individual household data from Luxembourg to evaluate how severe economic conditions could affect bank exposure to the household sector. Using data from a representative survey, information on household income, expenses and liquid assets are used to calculate a household-specific probability of default (PD). Aggregate bank exposure at default (EAD) is obtained by multiplying these household-level PDs by their corresponding volume of outstanding loans and summing across the population of households. Aggregate bank loss given default (LGD) is calculated by assuming that banks recover real estate assets from defaulting households and liquidate them with a haircut. To simulate adverse economic conditions, the exercise is repeated with scenarios combining severe but plausible shocks (i.e. tail risk) to real estate prices, bonds and stocks, household income and interest rates. Compared to the no-shock baseline, the LGD rises by a multiple of eight, reaching 4.2% of total bank exposure to the household sector. Thus, bank losses appear to be quite sensitive to severe stress. The high-stress scenario also generates a relatively high percentage of defaults among socioeconomically disadvantaged households (i.e. low net wealth, low income, low education, three or more dependent children). For instance, households in the lowest income quintile see their PD rise from 9.3% in the no-shock baseline to 14.8% in the most severe scenario. Our main conclusion is that bank losses appear to be quite sensitive to financial stress, despite three mitigating factors in Luxembourg: indebted households tend to hold liquid assets that can help smooth shocks, household leverage tends to decline rapidly once mortgages have been serviced several years, and loan-to-value ratios at origination appear not to be excessive.
JEL-codes: D10, D14, E44, G01, G21
Keywords: Stress testing; Financial stability; HFCS; Household finance

The paper analyses the place held by Luxembourg in global value chains (GVC) by relying on trade in value added data retrieved from OECD inter-country input-output tables, available over the period 1995-2011. The analysis is multifaceted as the role of Luxembourg in GVC is analyzed across 50 advanced and emerging market economies, at the country level, at the sector level and over time. Results show that Luxembourg acts as an important chain-link in GVC as evidenced by its strong upstream and downstream interconnections with other partner countries. Luxembourg is primarily a buyer of foreign value added and less a seller of domestic value added. The major part of Luxembourg’s GVC trading partners is located in Western Europe suggesting that the supply chain network is not global for Luxembourg but rather regional. Notwithstanding this, the share of East Asian and Eastern European emerging countries - albeit relatively low compared to advanced economies - is increasing over the period of analysis. A similar observation prevails for the geographical breakdown of the origin (destination) of foreign (domestic) value added for domestic (foreign) final demand at the end of the value chain. The analysis unveils that Luxembourg possesses a comparative advantage in GVC in the finance and insurance industry. It is from the latter sector that the country retrieves the most important share of value added from GVC participation.

Keywords: International trade, Global value chains, Gross trade statistics, Trade in value added statistics, Inter-country input-output tables

JEL codes: D57, F14, F20, F21, F23

This report presents the methodology and main descriptive results of the second wave of the Cross-border Household Finance and Consumption Survey (XB-HFCS) conducted in 2014. The survey provides novel information on the economic and financial situation of households employed in Luxembourg but living in neighbouring countries (cross-border commuters), who contribute substantially to Luxembourg’s economy. We present results on the composition of their assets and liabilities, net wealth, income and consumption. Household net wealth of crossborder commuters is more equally distributed compared to that of employed households resident in Luxembourg. In addition, cross-border commuters have a higher median net wealth and gross income compared to those of the employed population in their country of residence. About 26% of their financial assets and 19% of their liabilities are located in Luxembourg. While the majority of the non-durable expenditures are done in the country of residence, cross-border commuters consume about 20% of their household income in Luxembourg.

Keywords: cross-border commuters, households, survey, assets, liabilities, wealth, income,consumption

JEL-Codes: D31, D14, C81, C83, J61

JEL Classication: C1, E5, F3, G1

Keywords: financial stability; systemic risk; macro-prudential policy; dynamic copulas; value at risk; shapley values; risk spillovers.

This paper investigates the interaction between residential housing prices and mortgage credit in Luxembourg over the period 1980Q1-2017Q1. We use a vector error correction framework to model this interaction and allow for feedback effects between the two variables. In the long-run, higher housing prices lead to a mortgage credit expansion, which in turn puts upward pressure on prices. The growing demand for mortgage credit is also sustained by positive net migration to Luxembourg. Construction activity is another important determinant of housing prices, in line with existing supply-side limitations on dwelling availability. These dynamics lead to a structural imbalance between housing supply and demand, with the latter being fueled by demographic factors, tax incentives and fiscal subsidies, as well as the low interest rate environment. While price dynamics are partially explained by these structural factors, our results suggest that over the last few years residential housing prices have been characterized by a moderate, but persistent, overvaluation with respect to market fundamentals. Between 2012Q1 and 2017Q1, the average overvaluation is estimated at 6.85% but its trend is decreasing in the last quarters. Results also show that housing prices have a slow rate of adjustment to deviations from fundamentals (only 2.2% of the misalignment is corrected each quarter) and they do not directly adjust to disequilibria in the mortgage market. These findings are supported by impulse response analysis, which suggests that shocks to the endogenous variables lead to permanent increases in housing prices.

Keywords: residential real estate, housing market, VECM, property price valuation

Les institutions financières ont recours au collatéral afin d'accroître leur financement, fournissant des actifs servant de garantie en cas de défaut sur leurs obligations. Cela est le cas pour les activités bancaires non traditionnelles dans lesquelles des accords de pension sont employés comme principale source de financement. Ceci est également le cas pour les banques commerciales qui emploient le collatéral dans le cadre de la gestion de change et, plus récemment, de la politique monétaire non conventionnelle.

Alors qu'une littérature abondante sur des modèles d’équilibre général stochastiques dynamiques (EGSD/DSGE[1]) a souligné le rôle du collatéral des emprunteurs dans l'amplification des chocs macroéconomiques, le rôle du collatéral utilisé par les banques pour accroître leur financement a généralement été négligé.

Un modèle quantitatif d'équilibre général a été adopté pour explorer dans quelle mesure l'amplification du cycle économique est affectée par l'utilisation du collatéral lors de l’établissement de contrats de dépôts entre les ménages et les banques. Dans le modèle, la capacité des banques à assurer l'intermédiation de fonds entre les épargnants et les emprunteurs dépend de la composition des différents actifs qu'elles sont en mesure d'apporter comme collatéral lors de leur refinancement, c'est-à-dire un actif réel sécurisé, en l’occurrence le capital, et un actif financier plutôt risqué, tels que les prêts accordés à la clientèle.

Les résultats issus du modèle montrent que les deux types d’actifs contribuent à relâcher la contrainte financière des banques. Cependant, il s’avère que l’augmentation du capital exacerbe la mauvaise allocation des ressources et engendre une réduction des prêts à l’économie, tandis que l’augmentation des actifs bilantaires contribue à l’amenuisement de l'écart de taux (spread) entre les prêts et les dépôts, atténuant ainsi la mauvaise allocation du capital.

De plus, le modèle reproduit le processus contracyclique d’une fuite en faveur des actifs de qualité dans l'allocation optimale des actifs bancaires. Comme résultat, les banques augmentent (diminuent) leur détention des actifs risqués durant les périodes d'expansion (récession), réduisant (augmentant) leur détention des actifs liquides. En conséquence, le levier bancaire est procyclique et génère des fluctuations excessives du crédit, de la production et des prix des actifs.

Du point de vue normatif, nous étudions dans quelle mesure un régulateur bancaire peut intervenir pour lisser l'amplitude de ces fluctuations en réduisant le mécanisme de propagation endogène qui repose sur la mauvaise allocation du capital.

Les résultats montrent qu'un ratio constant du capital sur actifs atténue la transmission d’un choc technologique, même si l'écart entre les produits marginaux du capital des emprunteurs et des banques ne peut être comblé entièrement. Par contre, le régulateur peut atténuer avec succès la réponse de l'économie à un choc technologique au travers de la fixation d'un coussin de fonds propres contracycliques, lequel contribuera à stabiliser les fluctuations du collatéral des emprunteurs, même si la distorsion de l'allocation du capital demeure non résolue.

Le contenu de cette étude ne doit pas être perçu comme étant représentatif des opinions de la Banque centrale du Luxembourg ou de l’Eurosystème. Les opinions exprimées reflètent celles des auteurs et non pas nécessairement la position de la Banque centrale, de ses dirigeants ou de l’Eurosystème.

 [1] Dynamic Stochastic General Equilibrium.

L’émergence de monnaies virtuelles, dont la plus célèbre est le bitcoin, a attiré beaucoup d’attention au cours des dernières années.[1] Le bitcoin diffère d’une monnaie nationale ayant cours légal et n’est pas régulé par une autorité monétaire.[2] Un algorithme informatique détermine l’émission de bitcoins, qui se produit à un rythme décroissant. Le nombre de bitcoins émis est réduit de moitié environ tous les quatre ans et le total des bitcoins en circulation ne dépassera jamais 21 millions. Les utilisateurs n’ont pas besoin de passer par une institution financière lorsqu’ils vendent ou achètent des articles avec des bitcoins. Les transactions avec des bitcoins sont traitées par des agents privés, les mineurs, qui sont rémunérés avec les frais de transaction et les bitcoins nouvellement émis. Au fur et à mesure que l’émission de monnaie virtuelle décroît, les frais de transaction devront augmenter pour que les mineurs continuent à fournir les services de paiement liés aux opérations avec les bitcoins. 

Cette étude analyse les conséquences d’une diminution de l’émission de monnaie virtuelle et de l’augmentation des frais de transaction qui s’en suit. Elle s’appuie sur un modèle macroéconomique standard en y incorporant (i) un bien supplémentaire, dit ‘virtuel’, vendu en monnaie virtuelle et (ii) des mineurs. Les biens virtuels résultent de la combinaison de biens intermédiaires et de services de paiement fournis par les mineurs. Les mineurs valident les transactions en monnaie virtuelle et sont payés avec de la monnaie virtuelle nouvellement créée et avec des frais de transaction, inclus dans les opérations avec la monnaie virtuelle. L’approche adoptée ici est que les consommateurs achètent les biens virtuels sur une plate-forme associant les biens physiques (intermédiaires) produits par les entreprises intermédiaires aux services de paiement fournis par des mineurs. 

L’analyse se concentre sur les conséquences d’une baisse de la croissance monétaire telle que prévue dans un système de monnaie virtuelle, comme le Bitcoin, où le taux d’émission de nouvelles pièces est initialement élevé et diminue progressivement jusqu’à atteindre zéro.

La principale conclusion de cette étude est que la croissance de la monnaie virtuelle, lorsque les mineurs sont payés avec de la monnaie virtuelle nouvellement créée, peut avoir des effets opposés à ceux prédits par la théorie monétaire standard. Une baisse du taux d’émission de la monnaie virtuelle agit à travers deux canaux. Un premier effet passe par une réduction de la taxe d’inflation émanant de la monnaie virtuelle, ce qui stimule la demande de biens virtuels (sans affecter les prix relatifs des biens). C’est le seul canal par lequel le taux d’émission affecte l’économie lorsque les mineurs sont uniquement rémunérés avec les frais de transaction. Cet effet standard passant par la taxe d’inflation est contrecarré par le mécanisme de récompense, qui apparaît lorsque les mineurs sont également rémunérés avec de la monnaie virtuelle nouvellement émise. A travers ce nouveau mécanisme, une baisse de l’émission de monnaie virtuelle augmente les frais de transaction, ce qui résulte en une hausse du prix relatif des biens virtuels (par rapport aux prix des autres biens de consommation) et en une réduction de leur demande. Des extensions du modèle considèrent les effets de taxes de consommation ainsi que d’une sécurité imparfaite du système de paiement virtuel (la sécurité s’entend comme la proportion d’encaisses de monnaie virtuelle garanties). Ces éléments - taxes de consommation et sécurité imparfaite - renforcent le mécanisme de récompense qui peut dominer l’effet traditionnel lié à la taxe d’inflation. 

Un autre résultat majeur est que les effets de l’émission de monnaie virtuelle sur le bien-être dépendent de la croissance de la monnaie nationale, c’est-à-dire la monnaie émise par l’autorité monétaire nationale. Le bien-être affiche une forme de cloche lorsque l’émission de monnaie virtuelle décroît: il s’améliore d’abord à des taux d’émission élevés, mais se détériore lorsque cette croissance atteint des taux faibles. Plus la croissance de la monnaie nationale est élevée, moins la baisse de croissance de la monnaie virtuelle sera bénéfique pour le bien-être. Enfin, le modèle est calibré et simulé pour des systèmes de paiement virtuel différents (la proportion de monnaie virtuelle nouvellement émise rémunérant les mineurs varie selon les systèmes). 

Le contenu de cette étude ne doit pas être perçu comme étant représentatif des opinions de la Banque centrale du Luxembourg ou de l’Eurosystème. Les opinions exprimées reflètent celles des auteurs et non pas nécessairement la position de la Banque centrale, de ses dirigeants ou de l’Eurosystème.

[1] Les monnaies virtuelles appartiennent à la famille des monnaies numériques, incluant toute monnaie stockée et transférée électroniquement. Il n’existe pas de définition unique de « monnaie virtuelle », car ces systèmes de paiement évoluent continuellement (ECB, 2012; IMF, 2016). Dans cette étude, le terme « monnaie virtuelle » désigne toute monnaie numérique acceptée comme paiement pour des biens réels, comme les cryptomonnaies (par exemple le bitcoin). En outre, ce document adopte la convention utilisant un B majuscule (Bitcoin) pour se référer au système de paiement et un b minuscule pour se référer à l’unité de compte (bitcoin). Finalement, qualifier les monnaies virtuelles de « monnaie » est sujet à débat. Certains auteurs soutiennent, en effet, qu’elles ne remplissent qu’imparfaitement les fonctions principales d’une monnaie: unité de compte, réserve de valeur et intermédiaire des échanges (p.ex. Yermack, 2014). Dans le contexte de l’Union Européenne, les « monnaies virtuelles » ne sont d’ailleurs pas considérées comme des monnaies (voir la définition de monnaie dans l’article 2 de la directive 2014/62/UE) et certaines institutions, telles que l’Autorité bancaire européenne, les qualifient de « représentations numériques d’une valeur » (EBA, 2014, p.11, paragraphe 19).

[2] En principe, le cours légal s’applique à la monnaie en circulation sur un territoire et signifie que cette monnaie ne peut être refusée en règlement d’une dette. En zone euro, l’euro est la monnaie unique et les billets et pièces mis à disposition par les banques centrales de l’Eurosystème (monnaie fiduciaire) doivent, de manière générale, être acceptés comme moyen de paiement. Par contre, la monnaie scripturale, comme les dépôts bancaires dans des comptes courants, n’a pas de cours légal (p.ex. un créancier n’est pas tenu à accepter un règlement par transfert bancaire).

La récente crise financière mondiale et ses conséquences ont incontestablement révélé l'insuffisance du cadre de la régulation des intermédiaires financiers traditionnels et mis en lumière la complexité de certaines activités de l’industrie financière et parfois de leur l’opacité, qualifiée communément de “système bancaire ou système d’intermédiation parallèle”. Dans le même temps, les inquiétudes croissantes sur la vulnérabilité du système financier mondial à la suite de cette crise ont conduit les autorités au niveau mondial à concevoir une réponse réglementaire visant à endiguer les conséquences d’une capitalisation insuffisante et des pénuries de liquidité dans le système bancaire. Une telle réponse, connue sous le nom de Bâle III s’est manifestée par l’introduction de plus d’exigence en capital réglementaire et en liquidité pour les établissements de crédit et d'autres provisions devant être appliquées aux assureurs. En dépit de la nécessité de prendre des mesures nouvelles, la charge accrue de la conformité financière a soulevé de nouvelles inquiétudes pour les autorités de régulation puisqu'elles peuvent créer, à travers de nouvelles opportunités d’arbitrage réglementaire, des incitations additionnelles pour les établissements de crédit à transférer une partie de leurs activités en dehors de la sphère régulée, augmentant davantage la taille du secteur bancaire parallèle. Même si le rôle central du “système bancaire d’intermédiation parallèle” dans le boom des crédits au début des années 2000 et dans les faillites durant la récente crise financière de 2007-2009 a été largement étudié, le nombre d’études analysant ses implications sur l’économie réelle et les mécanismes de transmission sous-jacents est étonnamment limité. Cette étude contribue à combler ce vide et apporte un éclairage nouveau sur le rôle controversé joué par le système bancaire parallèle dans la transmission des chocs à l’aide d’un modèle néokeynesien d’équilibre général stochastique et dynamique (DSGE). 

Dans le cadre de la modélisation adoptée, les “intermédiaires parallèles” opèrent sur le marché interbancaire en prêtant des fonds aux banques commerciales, sur le marché traditionnel de crédits en fournissant des prêts aux grandes entreprises et sur le marché secondaire de prêts en achetant des “titres adossés à des actifs” émis par des banques commerciales. De manière cruciale, le comportement des banques commerciales est sujet au problème d’aléa moral induit par l’existence de multiples opportunités alternatives d’investissement, laquelle leur fournit l’incitation à libérer des capitaux propres en recourant à la titrisation “des créances” à leur cession aux “intermédiaires parallèles” sur le marché secondaire.

L’implication majeure est que n’importe quel transfert de risque du système bancaire traditionnel au “système parallèle” permis par la titrisation retourne au premier secteur à travers le marché interbancaire et dans le secteur productif à travers l’attribution de prêts aux entreprises. Les instruments macroprudentiels sont mis en œuvre dans l’objectif d'atténuer les effets indésirables induits par la titrisation. Ils comprennent le ratio de levier qui impose le maximum d’exposition des établissements de crédit vis-à-vis des petites entreprises pour un niveau donné de capitaux propres, et le ratio de titrisation qui impose une limite maximale de prêts pouvant être titrisés et cédés sur le marché secondaire. Les résultats montrent que la complémentarité de tels instruments permet aux autorités macroprudentielles d'atteindre, avec succès, la stabilisation macroéconomique à la suite d’un choc, puisque leur activation simultanée serait efficace pour réduire la volatilité du produit intérieur brut.

We construct debt burden indicators at the level of individual households and calculate the share of households that are financially vulnerable using Luxembourg survey data collected in 2010 and 2014. The share of households that were indebted declined from 58.3% in 2010 to 54.6% in 2014, but the median level of debt (among indebted households) increased by 22% to reach € 89,800. This suggests that indebted households in 2014 carried a heavier burden than indebted households in 2010. However, among several debt burden indicators considered, only the debt-to-income ratio and the loan-to-value ratio of the outstanding stock registered a statistically significant increase. The median debt service-to-income ratio actually declined, mainly reflecting lower costs on non-mortgage debt. Using conventional thresholds to identify financially vulnerable households, we find that their share in the population of indebted households increased, although the change was only statistically significant when measured by the debt-to-income ratio. The different indicators of debt burden and financial vulnerability are highly correlated with several socio-economic characteristics, including age, gross income and net wealth. In particular, low income households have lower leverage and disadvantaged socio-economic groups (in terms of education, employment status and homeownership status) tend to be less financially vulnerable. However, after controlling for other factors, low income or low wealth increase the probability of being identified as vulnerable.

The paper analyses gross portfolio investment flows in equity and investment fund shares (EIFS) in Luxembourg - a small open economy with a financial center - over the period 2002Q1-2016Q3. The statistical analysis shows that gross EIFS flows exhibit similar patterns over time amongst resident investors and non-resident investors. However, the volatility of EIFS flows instigated by non-resident investors is larger than the volatility of EIFS flows initiated by resident investors. The graphical analysis provides evidence that gross EIFS flows switch between positive and negative growth cycles whose durations vary over time, depending on macroeconomic, financial and geopolitical shocks at the global level. In particular, gross EIFS flows correlate positively with stock returns and negatively with risk/uncertainty measures at the global level. Sudden and sharp increases (decreases) in gross EIFS flows concur with periods of bullish (bearish) equity markets and low (heightened) risk aversion. Econometric tests show that gross EIFS flows (including extreme movements) are driven by macroeconomic and financial variables at the global level. Eventually, a prediction exercise suggests that it is difficult to forecast extreme movements in gross EIFS flows based on global macroeconomic and financial variables.

Keywords: International finance, external statistics, balance of payments, equity and investment fund shares, gross flows, surges/flights, stops/retrenchments, graphical analysis, GMM estimation, discrete choice model, ROC analysis, prediction exercise

In this paper, we revisit the role of regulation in a small-scale dynamic stochastic general equilibrium (DSGE) model with interacting traditional and shadow banks. We estimate the model on US data and we show that shadow banking interferes with macro-prudential policies. More precisely, asymmetric regulation causes a leak towards shadow banking which weakens the expected stabilizing effect. A counterfactual experiment shows that a regulation of the whole banking sector would have reduced investment fluctuations by 10% between 2005 and 2015. Our results therefore suggest to base  regulation on the economic functions of financial institutions rather than on their legal forms.

KEYWORDS: Shadow Banking, DSGE models, Macro-prudential Policy.
JEL CLASS.: C32, E32.

We analyse the use of active labour market policy (ALMP) measures and short-time work arrangements (STWAs) by Luxembourg firms during the years of economic and financial crisis (2008-09) and the subsequent European sovereign debt crisis (2010-13). About 34% of Luxembourg firms used ALMPs between 2008 and 2013. Economy-wide, use of ALMPs increased along both the extensive margin (more firms) and the intensive margin (more measures per firm). The likelihood that a firm hired with recourse to ALMPs is greater for large, domestically oriented, multiple establishment firms, firms facing strong demand, with concerns about labour cost pressures and unavailability of skilled labour. The crisis saw a surge in firms using STWAs. The likelihood of applying for STWAs increases with demand volatility, the share of workers with permanent contracts, export orientation and the inability to shift workers between establishments. Firms reported that 20-25% of jobs in STWAs were saved by this measure.

This study takes a comprehensive approach to systemic risk stemming from Luxembourg’s Other Systemically Important Institutions (OSIIs), from the Global Systemically Important Banks (G-SIBs) to which they belong, from the investment funds sponsored by the OSIIs, from the housing market, from the non-financial corporate sector and from the sovereign. All sectoral balance sheets are integrated and the resulting systemic contingent claims are linked into a stochastic version of the general government balance sheet to gauge their impact on sovereign risk. Explicitly modeling default dependence and capturing the time-varying non-linearities and feedback effects typical of financial markets, the approach evaluates systemic losses and potential public sector costs from contingent liabilities stemming directly or indirectly from the financial sector. Various vulnerability and risk indicators suggest the sovereign is robust to a variety of shocks. The analysis highlights the key role of a sustainable fiscal position for financial stability.

The purchase of securities, and more specifically government bonds, belongs to the monetary policy implementation framework of many central banks, the Eurosystem being no exception for that matter.

However, as for the euro zone, that tool remained unused until 2010, while present in the Eurosystem’s toolkit since its creation. Its implementation in times of crisis raised many debates, comments and even resorts to courts of justice. One of the central issues relates to the monetary financing of the public sector which in turn questions the relations between public debt, central banks and money.

This paper does not aim at providing a definite answer to the many questions, or to offer an arbitrage between the different arguments and schools of thoughts. More simply, in view of the often confused state of discussions, it goes back to the basic concepts of money creation, more specifically to the one of money creation by central banks for the benefit of the public sector.

Through a series of “typical cases” of interactions between central banks, commercial banks, public sector and households, the paper favours a better understanding of the quite complex mechanic of money creation through the purchase of public bonds by central and commercial banks. It also addresses a connected topic, i.e. the article 123 of the treaty on the Functioning of the European Union that prohibits the direct purchase by central bank on the primary market of debt instruments issued by the public sector.

This paper illustrates the main features of the Labour Module of the CompNet dataset which provides indicators of firm growth over the period 1995-2012 across 17 EU (13 euro area) countries and 9 macro-sectors. It also includes information on a large set of micro-aggregated characteristics of firms growing at different speed such as their financial position and labour and total factor productivity. The paper shows that during the Great Recession the share of shrinking firms sharply increased in countries under stress, while firm growth slowed down in non-stressed countries. In the former, the construction sector suffered the most, while in the latter manufacturing and services related to transportation and storage were mainly affected, possibly as a result of the trade collapse. While we find that, all else equal, more productive firms had a higher probability of growing, the process of productivity-enhancing reallocation was muted during the Great Recession.

This report presents the main results and the underlying methodology of the 2nd wave of the Luxembourg Household Finance and Consumption Survey (LU-HFCS) and compares them to those obtained in the 1st wave in 2010. This survey is conducted among private households resident in Luxembourg and is part of the Eurosystem Household Finance and Consumption Survey, which provides detailed individual and household data on assets, liabilities, income and consumption. This individual-level information on households provides a view on the distribution of assets and liabilities that complements the aggregate data on the household sector in the financial accounts.

This paper incorporates sticky investment prices in a two-sector monetary model of business cycles. Fit to quarterly U.S. time series, the model suggests that price rigidity in the investment sector is the single most empirically relevant friction to match the data. Sticky investment prices are also important to understand the dynamic effects of technology shocks and their pass-through to the relative price of investment goods.

This report presents new insights on the nature, size and persistence of various shocks (demand, credit, costs etc.) experienced by Luxembourg firms during the initial years of the economic crisis in 2008-09 and subsequently in the period 2010-13, as well as on how firms adjusted to these shocks in terms of employment, wages and prices. It discusses the extent to which institutional changes in the Luxembourg labour market through various public support measures helped alleviate the effects of the economic crisis.

Keywords: Economic and financial crisis, reaction to shocks, wage and price rigidity,

firms, survey, WDN

JEL Codes: C25, D22

This paper characterises the financial cycle in Luxembourg using both the growth and classical cycle definitions. We implement both a frequency-based approach –using band-pass filters– to measure the growth cycle and a turning-point approach to capture the classical cycle. The financial cycle is characterized using varibales related to domestic credit and asset prices. We identify the dates of peaks/troughs for growth and classical cycles, describe the characteristics
of cycle phases and analyze the synchronisation between cycles for each macro-financial variable considered and the real activity. Additionally, we evaluate the synchronisation of credit and house prices across the neighbouring countries, based on the medium-term classical cycle. Finally, we introduce two novel tools to monitor the evolution of the financial cycle which are intended to contribute to informing macroprudential policy. The first tool is an optimal decision rule in the form of two warning thresholds signalling growth cycle phases related to a possible classical turning-point. The second tool is a measure of the probability of a turning-point in the classical cycle in each quarter after a peak in the growth cycle. The tools are built on the lead/lag relationships between peaks and troughs of growth and classical cycles. A composite index of the growth cycle is proposed as well.
Keywords : financial cycles, turning-points,  synchronisation, band-pass filter, survival data, Area
Under the Receiver Operating Characteristic Curve, Luxembourg.
JEL classification : E32, G01, G18.

This study provides the first available estimates of systemic risk in the financial sector comprising the banking and investment fund industries during 2009Q4 -2015Q4.
Systemic risk is measured in three forms: as risk common to the financial sector; as contagion within the financial sector and; as the build-up of financial sector’s vulnerabilities over time, which may unravel in a disorderly manner. The methodology models the financial sector components’ default dependence statistically and captures the time-varying non-linearities and feedback effects typical of financial markets. In addition, the study estimates the common components of the financial sector’s default measures and by identifying the macro-financial variables most closely associated with them, it provides useful input into the formulation of macro-prudential policy. The main results suggest that: (1) interdependence in the financial sector decreased in the first three years of the sample, but rose again later coinciding with ECB’s references to increased search for yield in the financial sector. (2) Investment funds are a more important source of contagion to banks than the other way round, and this is more the case for European banking groups than for Luxembourg banks. (3) For tracking the growth of vulnerabilities over time, it is better to monitor the most vulnerable part of the financial sector because the common components of systemic risk measures tend to lead these measures.

JEL Classification: C1, E5, F3, G1

Keywords: financial stability; macro-prudential policy; banking sector; investment funds; default probability; non-linearities; generalized dynamic factor model; dynamic copulas.

Cette étude s’attache à évaluer dans quelle mesure la provision forfaitaire contribue à atténuer la procyclicité de la profitabilité bancaire et à déterminer l’impact de ce dispositif sur les recettes fiscales de l’Etat. Cette provision représente certes un manque à gagner en termes de recettes fiscales à court terme, mais celui-ci est provisoire étant donné qu’elle est incorporée au résultat de la banque et, in fine, taxée. Les résultats des estimations indiquent que la provision forfaitaire suit une évolution opposée aux cycles financier et réel lorsque ceux-ci sont approximés, respectivement, par les écarts à la tendance de long terme du prix de l’immobilier résidentiel et du PIB luxembourgeois. Elle contribue également à lisser le profit des banques et, par là même, elle lisse les recettes fiscales et contribue à la stabilisation du solde budgétaire à la suite d’un retournement du cycle. La provision forfaitaire permet donc de mieux couvrir les pertes attendues – insuffisamment couvertes par les provisions spécifiques seules – et limite l’absorption des fonds propres des banques lors des phases basses du cycle. En ce sens, elle peut s’avérer un complément au coussin de fonds propres contracyclique pour préserver la résilience du système bancaire.
Mots-clés : Provisions, Lissage du profit, Régulation bancaire.
Classification JEL : E32 ; G21 ; G28.

LOLA 2.0 is a dynamic general equilibrium model for the Luxembourg economy, which features overlapping generation dynamics, labormarket frictions à la Diamond-Mortensen- Pissarides and a New Open Economy Macroeconomics structure. This paper presents the model LOLA 3.0, which essentially integrates a financial sector to LOLA 2.0. In contrast to the existing dynamic stochastic general equilibrium (DSGE) literature, the financial sector does not intermediate between resident households and resident firms, but exports wealth management services. We calibrate the model to match the size of the financial sector in terms of employment, value added, net exports and taxes. The 2008 financial crisis has affected Luxembourg’s financial sector and slowed inflows of cross-border workers. Because there is a lot of uncertainty surrounding future growth of the Luxembourg financial sector and cross-border worker inflows, we use LOLA 3.0 to study the evolution of the Luxembourg economy between 2015 and 2060 under alternative scenarios (high – medium – low).

Keywords: Overlapping generations, Long-run projections, Financial sector, Luxembourg.

JEL-Code: D91, E24, E62, F41, J11.

After the household main residence (HMR), other real estate property (OREP) accounts for the second largest share of total household net wealth in the euro area. However, OREP remains mainly unstudied. Using Eurosystem HFCS data, I analyse OREP investment in Luxembourg, in selected euro area countries and in the euro area as a whole. For those households that own OREP, it represents an important share of gross wealth and generates non-negligible rental income. I identify several stylised facts of OREP characteristics with respect to their owners, type, use and location.

JEL Classification: D10, D14, D31

Keywords: household finance, individual portfolio choice, real assets

The primary aim of this work is to study the sensitivity of Luxembourg bond funds to interest rate movements. For this purpose, the dataset compiled at the Banque centrale du
Luxembourg (BCL) since December 2008 is used to analyse the balance sheet composition of Luxembourg bond funds and to measure the interest rate exposure of their
bond portfolio. An econometric model with time-varying parameters is then estimated on monthly data over the sample 2008:1-2014:6 to analyse the evolution of the interest rate
sensitivity of the Net Asset Value (NAV) of Luxembourg bond funds. The main findings of the study are the following. At the end of the period under review, Luxembourg bond
funds have lengthened the residual maturity and the duration of their portfolio, which have returned to a similar level as the one observed in December 2008. This evolution,
which points toward a search-for-yield behaviour in a low interest rate environment, suggests that Luxembourg bond funds have recently become more sensitive to fixedincome
market developments. According to the level of the parameter estimate obtained with the Kalman filter at the end of the sample, a 100 basis points rise in long term
interest rates on the sovereign bond markets associated with an additional 100 basis points rise in the risk premium on the high-yield bond markets would materialise
approximately into a 10% decrease in the NAV of Luxembourg bond funds.
Keywords: Bond funds, risk analysis, security-by-security reporting, Kalman filter.
JEL classification: C32, C81, F30, G11, G23.

This paper measures the links between financial services and other production sectors in the Luxembourg economy. The focus is on propagation within the country, without considering growth abroad. Among the 29 sectors in the annual input-output tables, financial intermediation is one of only four “key sectors” that the Leontief inverse identifies as featuring above-average forward and backward linkages to other production sectors. The links from financial services to other sectors became even stronger from 1995 to 2009. At quarterly frequency, Granger causality tests find that no sector leads financial services, although they also fail to find evidence that financial services lead other sectors. Lack of evidence may be attributed to quarter-on-quarter growth rates deviating from the normal distribution, with “fat tails” possibly reflecting volatility clustering. We therefore estimate univariate ARIMAGARCH models to identify normally distributed innovations in each sector. Comparing these growth innovations at different leads or lags, Cheung-Ng tests find little evidence of crosssector causality-in-mean or causality-in-variance. This time, lack of evidence could reflect time variation in cross-sector correlations, which is confirmed by the Engle-Sheppard test.
Estimated dynamic conditional correlations vary significantly through time, with cross-sector correlations surging during the financial crisis. Dynamic correlations are used to decompose the overall volatility of Luxembourg’s macroeconomic “portfolio” of different production sectors, which responds mostly to changes in cross-sector correlations. In conclusion, the financial sector appears to be strongly linked with the rest of the economy, benefiting from growth in other sectors, which it amplifies and propagates to the whole economy.
JEL classifications: C22; C51; C52; E0
Keywords: Output growth, GARCH models, Dynamic conditional correlations

This paper investigates the theoretical effects of immigration in an occupational choice model with three sectors: a low-skilled, a high-skilled and a public sector. The originality of our approach is to consider (i) intersectoral mobility of labor and (ii) public employment. We highlight the fact that including a public sector is crucial, since omitting it implies that low-skilled immigration unambiguously reduces wages and welfare of all workers. However, when public employment is considered, we demonstrate that immigration increases wages in the high-skilled and the public sectors, provided that the immigrant workforce is not too large and the access to public jobs is not too easy. The average wage of natives may also increase accordingly. Moreover, immigration may improve workers' welfare in each sector. Finally, the mechanism underlying these results does not require complementarity between natives and immigrants.
Keywords: Immigration, occupational choice model, public employment
JEL Classification: J24, J61, J45, H44

This study measures investment funds’ systemic credit risk in three forms: (1) credit risk common to all funds within each of the seven categories National Central Banks report
to the ECB; (2) credit risk in each category of investment fund conditional on distress on another category of investment fund and; (3) the build-up of investment funds’
vulnerabilities which may lead to a disorderly unraveling. The paper uses a novel framework which combines marginal probabilities of distress estimated from a structural
credit risk model with the consistent information multivariate density optimization (CIMDO) methodology and the generalized dynamic factor model (GDFM). The
framework models investment funds’ distress dependence explicitly and captures the time-varying non-linearities and feedback effects typical of financial markets. In addition,
the estimates of the common components of the investment funds’ distress measures may contain some early warning features, and identifying the macro and financial
variables most closely associated with them may serve to guide macro-prudential policy.
The relative importance of these variables differs from those associated with the common components of marginal measures of distress. Thus this framework can
contribute to the formulation of macro-prudential policy.

JEL Classification: C1, E5, F3, G1
Keywords: financial stability; investment funds; procyclicality, macro-prudential policy;
structural credit risk models; probability of distress; non-linearities; generalized dynamic
factor model; dynamic copulas.

Keywords: Household financial decisions, individual portfolio choice, real and financial assets, cross-country comparisons

JEL Classification: D1, D3

Jel -Classification: C8; D12; D14; D91

Keywords: Household Finance and Consumption; Life Cycle Saving; Survey Data

Since October 2008, the credit granted by the Eurosystem to the Euro zone banking sector increased in a substantial way, as a result of the implementation of nonconventional measures, in particular the fact that the Eurosystem left to the banks the faculty to determine themselves the quantity of credit that they wished to obtain.

This paper first recalls the foundations of the interbank money market and then analyses the evolution of the “net liquidity needs” of the banking sector. It provided a clarification of the relation between the Eurosystem, the euro zone banking sector and the money market. In particular, it develops arguments against the myth of “idle money parked with the Eurosystem”.

Keywords: monetary policy implementation, central bank, central bank’s balance sheet, money market, liquidity deficit, excess liquidity

Results from the Eurosystem Household Finance and Consumption Survey reveal substantial variation in household net wealth across euro area countries that await explanation. This paper focuses on three main factors for the wealth accumulation process, i) homeownership, ii) housing value appreciation and iii) intergenerational transfers. We show that these three factors, in addition to the common household and demographic factors, are relevant for the net wealth accumulation process in all euro area countries, and moreover that, using various decomposition techniques, differences therein, in particular in homeownership rates and house price dynamics, are important for explaining wealth differences across euro area countries.

Keywords: household wealth, homeownership, property prices, inheritance, euro area

JEL Codes: D31, E21, O52, C42

Crossing borders, be it international or regional, often go together with price, wage or indeed wealth discontinuities. This paper identifies substantial wealth differences between Luxembourg resident households and cross-border commuter households despite their similar incomes. The average (median) net wealth difference is estimated to be €367,000 (€129,000) and increases for higher percentiles. Using several different regression and decomposition techniques, spatial (regional) differences in real estate price developments, and thus differences in accumulated nominal capital gains are shown to be one main driving factor for these wealth differences. Other factors contributing to the observed wealth differences are differences in age, income, education and other household characteristics.

Keywords: household survey, wealth, real estate price dynamics, cross-border commuting

JEL Codes: D31, J61, F22, R23, R31

This paper analyses empirically how cross-border consumption varies across product and services categories and across household characteristics. It focuses on the part of crossborder sales that arise due to work-related cross-border crossings; it analyses the crossborder consumption behaviour of cross-border commuter households residing in Belgium, France and Germany and working in Luxembourg. In total, it is estimated that these households spend €925 million per annum in Luxembourg, reflecting about 17% of their gross annual income from Luxembourg and contributing about 10% to total household final consumption expenditure in Luxembourg. Cross-border consumption expenditure is shown to depend on individual and household characteristics, such as total household income, the number of cross-border commuters in the household, distance between home and work, as well as price level (index) differences between Luxembourg and its neighbouring countries.Cross-border commuters take advantage of existing arbitrage opportunities.

Keywords: cross-border shopping, commuting, consumption, expenditure, households

JEL Codes: F15, R12, R23, J61

In 2007, a new indicator of economic activity for Luxembourg was elaborated at the BcL. It was developed using a large database of about 100 economic and financial time series. The methodology was based on the generalized dynamic-factor models, and the model was estimated over the period from June 1995 to June 2007. Forecast performance was evaluated on several criteria (both in pseudo-real-time and using ex-post in-sample simulations) and results were satisfactory. They gave in particular clear evidence that the indicator provides better forecasts of GDP growth than a more standard approach that relies on past GDP values only. In this paper, we present results of the real-time use of the indicator from December 2007 onwards. Special attention is given to real-time forecasts of GDP growth and the real-time assessment of the economic situation that were made during the financial crisis. The root mean squared forecast errors of the indicator-based GDP growth forecasts have decreased during the 2009-2011 ''recovery" period in comparison to the 2007-2009 period, which is an encouraging result. This paper also includes results based on an extended study period, which includes (real-time) forecasts that were produced until the end of April 2013. The mean squared errors appear to have on average decreased over the second half of this extended study period in comparison with the first half. Finally, the BcL indicator produced better forecasts on average than the benchmarks over this extended study period.

JEL Classification:  C53, E17

Keywords:  Forecasting, factor model, large datasets, real-time analysis

Outre au problème de vieillissement de la population, le Luxembourg devra faire face au départ à la retraite de larges contingents de travailleurs non-résidents ainsi qu’à l’essoufflement attendu de l’immigration et de l’arrivée de futurs frontaliers. Le financement des systèmes de pensions par répartition devrait s’avérer plus difficile que dans d’autres pays.

 Nos précédentes études, basées sur le modèle macroéconomique LOLA, montrent que les changements démographiques vont sérieusement hypothéquer la santé des finances publiques et que la récente réforme des pensions - même si elle va dans la bonne direction - n’est pas suffisante pour régler le problème des pensions et maintenir des finances publiques saines.

Dans cette étude, nous proposons et évaluons une réforme globale, appelée LOLA. Cinq conclusions majeures ressortent de notre analyse. Premièrement, la “réforme LOLA” permet de contenir le problème des finances publiques à moyen terme (jusqu’en 2060) en maintenant le déficit public hors soins de santé aux alentours de 3%du PIB. Deuxièmement, elle préserve la croissance économique et limite la perte de bien-être des générations actuelles. Troisièmement, elle assure une plus grande équité intergénérationnelle que la récente réforme des pensions. Quatrièmement, ces conclusions se vérifient pour différents cas de figure - optimistes et pessimistes - d’évolution de la démographie et du travail frontalier. Finalement, nous montrons comment mettre en oeuvre la “réforme LOLA” en pratique et de façon complémentaire à la récente réforme.

Mots-clés : Générations imbriquées, Projections de long-terme, Pensions, Luxembourg.

JEL-Code : D91, E24, E62, F41, J11.

The aim of this work is to investigate the impact of the exchange rate on Luxembourg equity funds. For this purpose, the dataset compiled by the Banque centrale du Luxembourg is used to exploit the detailed information on the currency composition of assets and liabilities and to deliver a statistical decomposition of the exchange rate valuation effect on both sides of the balance sheet. In addition, an econometric analysis relying on the International Capital Asset Pricing Model (ICAPM) is carried out to estimate the sensitivity of the Net Asset Value (NAV) to exchange rate movements. The main findings of the study are the following. (i) Equity funds in Luxembourg are highly internationalized as for the currency composition of their balance sheet, with 54% of noneuro denominated shares on the liability side, and 81% of non-euro denominated securities on the asset side at the end of June 2013. (ii) The currency composition of Luxembourg equity funds has changed since the onset of the financial crisis, with relatively more USD-denominated shares on the liability side and relatively more emerging markets currency-denominated securities on the asset side. (iii) This structural change in the currency composition of Luxembourg equity funds delivered a higher sensitivity of the NAV to exchange rate movements, in particular with respect to the emerging markets currencies. (iv) Despite this increased sensitivity to the exchange rate, stock market developments remain the most important driver for the activity of Luxembourg equity funds in the medium run. From this point of view, the EUR/USD exchange rate provided a natural hedging against stock market fluctuations during the crisis period, thereby mitigating the aggregate evolution of the NAV expressed in euro.

Keywords: Equity funds, risk analysis, ICAPM, fixed-effects model.

JEL classification: F30, G11, G23.

This study investigates whether and how the crisis in 2008/2009 affects households' risk attitudes, subjective risk and return expectations, and planned financial risk taking using the German SAVE study. Households' wealth change from end-2007 to end-2009 is not found to have an effect. However, households that attribute losses to the crisis decreased their risk tolerance and planned risk taking; the probability of expecting an increase in risks and returns is raised. According to economic theory, wealth changes attributed to a dramatic event should not have a different effect than other wealth changes. The results suggest an emotional reaction.

Keywords: Financial and economic crisis, risk preferences, stock market expectations, wealth uctuations, emotions

JEL-Codes: D81, D14, G11

Mortgages constitute the largest part of household debt. An essential choice when taking out a mortgage is between fixed-interest-rate mortgages (FRMs) and adjustable-interest-rate mortgages (ARMs). However, so far, no comprehensive cross-country study has analyzed what determines household demand for mortgage types, a task that this paper takes up using new data for the euro area. Our results support the hypothesis of Campbell and Cocco (2003) that the decision is best described as one of household risk management: income volatility reduces the take-out of ARMs, while increasing duration and relative size of the mortgages increase it. Controlling for other supply factors through country fixed effects, loan pricing also matters, as expected, with ARMs becoming more attractive when yield spreads rise. The paper also conducts a simulation exercise to identify how the easing of monetary policy during the financial crisis affected mortgage holders. It shows that the resulting reduction in mortgage rates produced a substantial decline in debt burdens among mortgage-holding households, especially in countries where households have higher debt burdens and a larger share of ARMs, as well as for some disadvantaged groups of households, such as those with low income.

JEL-codes: D12, E43, E52, G21

Keywords: mortgage choice, fixed-rate mortgage, adjustable-rate mortgage, household finance, monetary policy.

We propose a fully flexible, complete-market model of the international business cycle that is consistent with two major empirical facts: positive cross-country co-movement of economic aggregates and a negative correlation between the real exchange rate and relative consumption (the Backus-Smith puzzle). The model features non-tradable goods, zero wealth effects on labour supply, imperfect substitutability of capital across sectors and variable capacity utilisation. The latter can generate strong Balassa-Samuelson effects that drive a low consumption-real exchange rate correlation. Cyclical movements across countries are also positively correlated. The novelty of our paper is to introduce changes in expectations (news-shocks) as an explanation to the Backus-Smith puzzle through the wealth effects of future changes in income, while being consistent with expectations-driven economic expansions.

Keywords: News-Driven Cycles, Backus-Smith Puzzle, Real-Exchange Rates

JEL Classiffication: F41, F44

This study proposes a novel framework which combines marginal probabilities of default estimated from a structural credit risk model with the consistent information multivariate density optimization (CIMDO) methodology of Segoviano, and the generalized dynamic factor model (GDFM) supplemented by a dynamic t-copula. The framework models banks’ default dependence explicitly and captures the time-varying non-linearities and feedback effects typical of financial markets. It measures banking systemic credit risk in three forms: (1) credit risk common to all banks; (2) credit risk in the banking system conditional on distress on a specific bank or combinations of banks and; (3) the buildup of banking system vulnerabilities over time which may unravel disorderly. In addition, the estimates of the common components of the banking sector short-term and conditional forward default measures contain early warning features, and the identification of their drivers is useful for macroprudential policy. Finally, the framework produces robust outof-sample forecasts of the banking systemic credit risk measures. This paper advances the agenda of making macroprudential policy operational.

JEL Classification: C30, E44, G1

Keywords: financial stability; procyclicality, macroprudential policy; credit risk; early warning indicators; default probability, non-linearities, generalized dynamic factor model; dynamic copulas; GARCH.

This work uses dynamic panel data methods to identify the determinants of the loan loss provisioning ratio in the Luxembourg banking sector from 1995 Q1 to 2011 Q4. The study is motivated by the theoretical framework assuming that both macroeconomic and microeconomic variables have a strong impact on the loans quality and quantity. The empirical results for Luxembourg confirm the findings of previous studies: both macroeconomic and bank-specific variables have a large impact on the development of the loan loss provisioning ratio. Indeed, the results show that GDP growth, house prices, ROA and the solvency ratio have a negative impact on the loan loss provisioning ratio, whereas the unemployment and interest rate increase the ratio.

JEL classification codes : G21, C23.

Keywords: loan loss provisioning ratio Luxembourg banking system, macroeconomic factors and specific banking factors, dynamic data panel methods.

This paper attempts to empirically identify the determinants of Luxembourgish banks’ reliance on short term funding. The emphasis lies on making the link to developments in the macroeconomic environment and the build up of systemic risk while institution-specific factors are being controlled for. The paper provides evidence for a close link between exuberant credit developments at the aggregate level and short term funding of banks. This finding supports the view that one possible channel for increasing vulnerabilities during a lending boom may run through increased reliance of banks on short term funding. When it comes to bank specific variables, bank size has an important effect on the tendency to contract short term funding. This result is in line with recent work on leverage procyclicality in the banking sector. The results also imply that currently discussed regulatory standards on the funding structure of banks could mitigate the build up of vulnerabilities.

We study how the Basel III regulations, namely the Capital-to-assets ratio (CAR), the Net Stable Funding Ratio (NSFR) and the Liquidity Coverage Ratio (LCR), are likely to impact the banks’ profitabilities (i.e. ROA), capital levels and default. We estimate historical series of the new Basel III regulations for a panel of Luxembourgish banks for a period covering 2003q2 to 2011q3. We econometrically investigate whether historical LCR and NSFR components as well as CAR positions are able to explain the variation in a measure of a bank’s default risk (proxied by Z-Score), and how these effects make their way through banks’ ROA and CAR. We find that the liquidity regulations induce a decrease in average probabilities of default. Conversely, while we find that the LCR has an insignificant impact on banks’ profitability, those banks with higher NSFR (through lower required stable funding, the NSFR denominator) are found to be more profitable. Additionally, we use a model of bank behavior to simulate the banks’ optimal adjustments of their balance sheets as if they had had to adhere to the regulations starting in 2003q2. Then we predict, using our preferred econometric model and based on the simulated data, the banks’Z-Score and ROA. The simulation exercise suggests that basically all banks would have seen a decrease in their default risk if they had previously adhered to Basel III.

Keywords: Basel III, bank default, Z-Score, profitability, ROA, GMM estimator, simulation, Luxembourg.

JEL classification: G21, G28.

Exceeding 40% of domestic employment cross-border commuters are extremely important to Luxembourg’s economy and labour market in general. This paper presents unique information on their income, wealth and consumption using representative survey data from cross-border commuter households to Luxembourg. The estimated average total net wealth of cross-border commuter households is about €240,000, which falls substantially short of comparable estimates for Luxembourg resident households exceeding €700,000. Cross-border commuters do not only receive money from but also leave money in Luxembourg. In terms of consumption expenditures, they spend on average more than €9,300 per year inside Luxembourg’s borders, representing about 15% of their total gross income and 17% of their gross income from Luxembourg.

Keywords: household, survey, wealth, income, consumption, cross-border, commuter

JEL Codes: D31, C81, C83, J61

In this article we study the interaction between leading macroeconomic indicators (industrial production, stock prices, consumer sentiment and real interest rates) and financial sector leverage in major European countries. We base our analysis on monthly, country-aggregated panel VAR models for the pre-crisis period January 2003 to August 2008, and the crisis period September 2008 to June 2011. We find little evidence for a relationship between macroeconomic variables and leverage in the pre-crisis period, with only real interest rates having a negative short-term impact on leverage growth. We find positive feedback loops between sentiment and stock prices as well as MFI assets in the pre-crisis period, and a positive impact of real interest rate changes on equity and asset growth. Thus, balance sheet expansions were driven by sentiment and stock prices, while real interest changes allowed MFIs to profit from higher spreads. During the crisis period (starting in September 2008), we observe a countercyclical impact from leverage on sentiment and stock prices, while sentiment and stock prices bear a pro-cyclical impact on leverage. In contrast to this, MFI leverage in Luxembourg is negatively impacted by stock prices, suggesting significant impacts from marking-to-market.We conclude that leverage drives expectations of financial instability (via e.g. default expectations), while sentiment and stock prices drive financial institutions’ investment decisions (via e.g. collateral value effects).

Keywords: leverage; macroeconomic conditions; Panel VAR; GMM estimation.

JEL classification: G21; G32; E32.

Beyond cyclical movements, it may be helpful to understand how structural forces or policies shape an economy in the longer term. With such remote horizons, it is crucial to base analysis on an appropriate tool. In this paper, we build an overlapping generation structure with New Open Economy Macroeconomics and labour market frictions à la Diamond-Mortensen-Pissarides. The main novelty over LOLA 1.0 is the integration of current account and exchange rate dynamics according to the New Open Economy Macroeconomics approach. We calibrate the model on Luxembourg data. By way of illustration, we study the interactions between expected demographic changes, labour market dynamics and public finance, and we look at the recently proposed policy responses.

Keywords: Overlapping generations, Long-run projections, Imbalances, Luxembourg.

JEL-Code: D91, E24, E62, F41, J11.

The estimation of banks’ marginal probabilities of default using structural credit risk models can be enriched incorporating macro-financial variables readily available to economic agents. By combining Delianedis and Geske’s model with a Generalized Dynamic Factor Model into a dynamic t-copula as a mechanism for obtaining banks’ dependence, this paper develops a framework that generates an early warning indicator and robust out-of-sample forecasts of banks’ probabilities of default. The database comprises both a set of Luxembourg banks and the European banking groups to which they belong. The main results of this study are, first, that the common component of the forward probability of banks’ defaulting on their long-term debt, conditional on not defaulting on their short-term debt, contains a significant early warning feature of interest for an operational macroprudential framework driven by economic activity, credit and interbank activity. Second, incorporating the common and the idiosyncratic components of macro-financial variables improves the analytical features and the out-of-sample forecasting performance of the framework proposed.

JEL Classification: C30, E44, G1

Keywords: financial stability; macroprudential policy; credit risk; early warning indicators;default probability, Generalized Dynamic Factor Model; dynamic copulas; GARCH.

JEL: D12, D91, D14, J26

Key words: private pension, Riester, termination, financial literacy, SAVE

Keywords: Household, survey, editing, multiple imputation, wealth, income

JEL Codes: D31, D14, C81, C83

Keywords: financial social accounting matrix, computable general equilibrium models, financial accounts, portfolio choice, financial institutional sectors

JEL Classification: D30, C68, D57, G11, G20, D53

This paper estimates the contribution of financial shocks to fluctuations in the real economy by augmenting the standard macroeconomic vector autoregression (VAR) with five financial variables (real stock prices, real house prices, term spread, loans-to-GDP ratio and loans-to-deposits ratio). This VAR is estimated separately for 19 industrialised countries over 1980Q1-2010Q4 using three alternative measures of economic activity: GDP, private consumption or total investment. Financial shocks are identified by imposing a recursive structure (Choleski decomposition). Several results stand out. First, the effect of financial shocks on the real economy is fairly heterogeneous across countries, confirming previous findings in the literature. Second, the five financial shocks provide a surprisingly large contribution to explaining real fluctuations (33% of GDP variance at the 3-year horizon on average across countries) exceeding the contribution from monetary policy shocks. Third, the most important source of real  fluctuations appears to be shocks to asset prices (real stock prices account for 12% of GDP variance and real house prices for 9%). Shocks to the term spread or to leverage (credit-to-GDP ratio or loans-to-deposits ratio) each contribute an additional 3-4% of GDP variance. Fourth, the combined contribution of the five financial shocks is usually higher for fluctuations in investment than in private consumption. Fifth, historical decompositions indicate that financial shocks provide much more important contributions to output fluctuations during episodes associated with financial imbalances (both booms and busts). This suggests possible time-variation or non-linearities in macrofinancial linkages that are left for future research.

Keywords: Economic and financial crisis, reaction to shocks, wage rigidity

JEL Codes: C25, D22

Keywords: demographics; capital flows; overlapping generations; general equilibrium; unemployment

JEL Classification: J11; F21; D91; C68

JEL classification: G15, G21; G28, C14

Keywords: Foreign bank efficiency, Home-host country characteristics, Bank regulation, Data Envelopment Analysis, Bootstrap.

JEL Classification: C14; G21, D30

Keywords: Luxembourg Banks; Banking activities; Convergence; Distribution Dynamics; Non-Parametric kernel

JEL classification: E51, E52, E58, G21, G28.

Keywords: leverge dynamics, banking sector, GMM estimation, crisis effect.

JEL Classification: C30, E44, G1,

Keywords: financial stability; credit risk; structured products; default probability, GARCH.

Keywords: on-the-job search, cyclical properties

JEL classification: E24, E32, J64

JEL classification: C15, E44, G01, G21

Keywords: financial stability, stress testing, MVAR, mixture of normals, VAR, tier 1 capital ratio, counterparty risk, Luxembourg banking sector

Keywords: Overlapping Generations, Search Unemployment, Labor Force Participation, aging, Pensions, Labor Market

JEL-Code: E24, H55, J26, J64

In this paper we study the impact of the Basel III liquidity regulations, namely the Liquidity Coverage Ratio (LCR) and the Net Stable Funding Ratio (NSFR), on the bank lending channel in Luxembourg. For this aim we built, based on individual bank data, timeseries of the LCR and NSFR for a sample of banks covering between 82% and 100% of total assets of the banking sector. Additionally, we simulated the optimal balance sheet adjustments needed to adhere to the regulations. We extend the existing literature on the identification of the bank lending channel by adding as banks characteristics the estimated shortfalls in both the LCR and NSFR. We find a significant role for the bank lending channel in Luxembourg which mainly works through small banks with a large shortfall in the NSFR. We also show that big banks are able to increase their lending following a contractionary monetary policy shock, in line with the fact that big banks in Luxembourg are liquidity providers. Our extrapolation and simulation results   suggest that the bank lending channel will no longer be effective in Luxembourg once banks adhere to the Basel III liquidity regulations. We find that adhering to the NSFR may reduce the bank lending channel more strongly than complying with the LCR.

JEL classification: E51, E52, E58, G21, G28.

Keywords: bank, bank lending channel, monetary policy, Basel III, LCR, NSFR.

Keywords: subprime credit; randomized trials; liquidity constraints.

JEL Classi…cation: D11, D12, D14

JEL classification: C5, D14.

Keywords: Personal Finance, Financial Strain, Debt, Behaviour, Financial Literacy.

Keywords: household indebtedness, financial vulnerability, micro survey data, monetary transmission.

JEL Classification: C42, D12, D14, G21

Keywords: household, survey data, wealth gap, immigrants, distribution

JEL Codes: D31, F22

Keywords: wealth mobility, wealth dynamics, life cycle, heterogeneity, panel data

EconLit subject descriptors: C230, D140, D310, D910, H240

JEL Codes: D31, J15

Keywords: Inheritance, household wealth, inequality

Yet, individuals with low financial literacy are more likely to sell their assets which lost in value (realize losses). This reaction to short-term losses has potential longterm consequences if individuals do not participate in markets' recovery and face lower returns in the long run.

Keywords: financial literacy, cognitive ability, financial crisis, life-cycle savings, saving behavior, portfolio choice

JEL Classification: D91, D14, G11

JEL Classification: H31, E62, C83

Keywords: Fiscal stimulus, tax rebates, marginal propensity to consume, survey responses

JEL Codes: D91, E91 

Keywords: Credit constraints, mortgages, household consumption.

Keywords: Financial Advice, Trust, Consumer Protection, Household Finance

JEL codes: E1, G2, D8

Keywords: Access to finance, Bank-ownership, Deposit insurance, Payment system, Creditor protection.

JEL Codes: G2, G18, O16, P34

Keywords: Monetary policy, small open economy, VAR, macroeconomic shocks.

JEL classification: C32, E52, F41.

We report relatively small differences in the frequency of nominal wage cuts across occupational groups and sectors. In addition, the observed rigidity does not seem to be driven predominantly by the absence of negative shocks. We show that the downward real wage rigidity is related to automatic wage indexation, while additional factors might be necessary to explain the high degree of downward nominal wage rigidity.

JEL Code: J31.

Keywords: downward wage rigidity, wage indexation.

We use data on loan loss provisions and total loans over the period spanning 1995 until 2009 to estimate a stress testing model for the Luxembourg banking sector. The sample encompasses the recent global crisis and covers a period in which the average probability of default of the Luxembourg banking sector’s counterparties is observed to increase significantly. A joint model, consisting of several macroeconomic variables and the logit-transformed probability of default, is specified and estimated via seemingly unrelated regression (SUR). The results suggest that counterparty default rates are significantly affected by the euro area real GDP growth rate, the real interest rate and a domestic property price index. Conversely, changes in the Luxembourg real GDP growth rate have a much smaller effect on counterparty risk. We attribute this to the large number of foreign subsidiaries operating within Luxembourg. The estimated model is then used to simulate values of the probability of default and the macroeconomic variables over a horizon of 10 quarters. This allows us to construct distributions for the probability of default under both baseline and adverse scenarios. From the results of these simulations stressed Basel II tier 1 capital ratios are calculated and compared to their associated unstressed capitalization levels. Our calculations suggest that, under all the given adverse macroeconomic scenarios, the aggregate Luxembourg financial sector remains above the 4% minimum Basel II tier 1 capital requirement. Repeating the exercise on a limited sample of 5 individual banks produces similar results.

JEL classification: C15, E44, G01, G21

Keywords: financial stability; stress testing; Luxembourg banking sector; tier 1 capital ratio; counterparty risk

JEL Classification: G21, G28, G11

Keywords: Capital requirements, Credit risk, Deposit insurance, Prudential regulation, Portfolio approach. Published in the Journal of Financial Stability (Vol. 6, Issue 2, June 2010, pp. 79-84; doi:10.1016/j.jfs.2009.06.001)

In general, banks’ business lines and, therefore their buffers’ composition, determine the net effect of the shocks on banks’ stochastic liquidity buffers. So, results differ across banks. Second-round effects exemplify the relevance of contagion effects that reduce the systemic benefits of diversification. While systemic liquidity risk is low following a shock to the interbank market, for Luxembourg, with its high number of subsidiaries of large foreign financial institutions, the results indicate the importance of monitoring the liquidity of parent groups to which Luxembourg institutions belong. In particular, shocks to related-party deposits are important. Finally, the results, including those of a run-on-deposits shock, show the relevance of system-wide measures to minimize the systemic effects of liquidity crises.

JEL Classification Numbers: E5, C1, G2

Keywords: stress test, liquidity risk, banks, stochastic, contagion, macro-prudential

The implications of central bank collateral requirements for the monetary policy transmission mechanism and the working of the money market have often been neglected. Such implications, however, have clearly manifested during the course of the 2007-2009 crisis. As liquidity was vanishing in the interbank market, banks used (and abused of) central bank funding more intensively, in order to cover their financing needs. As a result, central bank collateral eligibility criteria have become even more critical than the policy rate as a factor of monetary policy transmission as well as a driver of liquidity in the interbank market. Thus, in the light of a retrospective analysis of some major events affecting monetary policy in the Euro area in the last two years, this study intends to properly reformulate the problem related to the choice of ‘optimal’ collateral requirements, by illustrating their interrelations with other central bank policy tools and targets. Ultimately, this approach allows us to derive a non-exhaustive set of recommendations for collateral and interest rate policies, as well as for central banks’ exit strategies from the current unconventional measures.

JEL Classification: E58, E51, E52

Keywords: Collateral, Monetary policy, Central bank, Interbank market, Financial stability

Keywords: Liquidity risk, Stress-test, Banking sector, Prudential supervision, Scoring system, Off-site supervision

JEL classification: G01, G21

Keywords: liquidity risk, liquidity stress testing, contingency funding plan, principal component analysis.

JEL Classification: G21

Keywords: Liquidity risk, Scenario analysis, Banking sector, Stress testing.

JEL classification: G21

JEL Classification System: E31,E32,E24,J64

Keywords: Inflation Dynamics, Labour Market, Business Cycle, Real Rigidities.

Keywords: Market potential, market access, regional wages, distance, European Union

JEL-Codes: F12, F15, R11, R12.

Keywords: Overlapping Generations, Search Unemployment, Small open economy, Labor Force Participation, Ageing, Labor Market Policy and Institutions

JEL-Code: E24, H55, J26, J64

Keywords: DSGE, Banking sector, Default risk, Supervision, Money

JEL classification: E13, E20, G21, G28

We build a model with frictional unemployment and staggered wage bargaining and we assume that hours worked are negotiated every period. We analyze the role of workers’ bargaining power in the hours negotiation on unemployment volatility and inflation persistence. The closer to zero is this parameter, (i ) the more firms adjust on the intensive margin, reducing employment volatility, (ii ) the lower the effective workers’ bargaining power for wages and (iii ) the more important is the hourly wage in the marginal cost determination. Combining staggered wage bargaining with some degree of workers’ bargaining power in the hours negotiation, we produce realistic labor market statistics together with inflation persistence.

Keywords: DSGE, Search and Matching, Nominal Wage Rigidity, Monetary Policy.

JEL classification: E31, E32, E52, J64.

La littérature économique récente a contribué à une meilleure connaissance des mécanismes de transmission de la politique monétaire aux taux des banques. Un certain nombre de ces études a permis de montrer que le niveau des taux d’intérêt, tout autant que le degré et la vitesse de transmission des conditions de financement, était très hétérogène au sein de la zone euro. Par conséquent, nous avons souhaité analyser ces éléments dans le cadre des banques luxembourgeoises, et, partant, nous prononcer sur une éventuelle spécificité des données luxembourgeoises par rapport à celles d’autres pays membres de la zone euro. Pour ce faire, nous avons conduit une analyse portant sur les crédits immobiliers, les crédits à la consommation et les dépôts tant au niveau agrégé qu’au niveau individuel. La période d’observation s’étend de 1993 à 2007 (données trimestrielles). Il est apparu que les résultats sont plutôt conformes aux attentes théoriques, en ce sens que la formation des taux débiteurs et créditeurs peut être expliquée par les conditions de financement. Par ailleurs, les degrés de transmission des taux de référence du marché aux différents taux de détail des banques luxembourgeoises se situent dans la médiane de la zone euro. De manière plus surprenante, nous savons que la spécialisation de la place dans les activités de banque privée joue un rôle dans l’hétérogénéité des données individuelles, mais ce point est difficile à mettre en lumière en terme statistique.

Keywords : Banks ; Depositary Institutions ; Micro Finance Institutions ; Morgages

JEL-Code : G21

Trotz weiterführender ökonomischer und politischer Integration ist die Europäische Union immer noch durch sehr niedrige Personenmobilität gekennzeichnet. Die regionale Arbeitsmobilität in der Europäischen Union wird in der Tat zum Teil so niedrig betrachtet, als dass sie nicht als ökonomisch relevanter Anpassungsmechanismus für regionale Arbeitsmarktungleichgewichte angesehen wird − ein Sachverhalt der auch für das Eurosystem von ausgewiesener Bedeutung ist. Dabei ist es gerade dieser regionale Aspekt, der wichtig erscheint, denn die regionalen Arbeitsmärkte in der Europäischen Union zeichnen sich durch starke Nachbarschaftseffekte aus. Das heißt, die regionale Arbeitsmarktsituation ist sehr eng mit der jeweiligen Arbeitsmarktsituation der benachbarten Regionen verknüpft. Regionale Mobilität kann jedoch nicht nur durch regionale Migration erreicht werden, sondern auch durch regionale Pendlertätigkeit. Und es scheint in der Tat so zu sein, dass diese zweite Form der regionalen Mobilität auf dem Vormarsch in der Europäischen Union ist. Laut Europäischer Kommission pendeln jeden Tag im Durchschnitt ungefähr 8 Prozent der Erwerbstätigen zwischen verschiedenen Regionen.

Das Ziel der vorliegenden Studie ist es, neue empirische Erkenntnisse über regionale Pendlerströme und seine Determinanten zu gewinnen. Ein wichtiger Aspekt ist in diesem Zusammenhang, ob regionale Pendlerströme auf regionale Arbeitsmarktungleichgewichte in der zu erwartenden Weise reagieren. Agieren Pendlerströme als ausgleichende Kraft, indem sie den Arbeitskräfteüberschuß in wirtschaftlich schwächelnden Regionen und den Arbeitskräftemangel in wirtschaftlich florierenden Regionen reduzieren? Anders ausgedrückt, können regionale Pendlerströme als potentieller Anpassungsmechanismus für regionale Arbeitsmarktungleichgewichte gesehen werden?

Die Resultate zeigen, dass Pendlerströme mit den Größen der Herkunft– und Zielregionen zunehmen und mit ihrer Entfernung von einander abnimmt. Die empirischen Schätzergebnisse belegen weiterhin, dass die Pendlerströme in der Tat auf Arbeitsmarktungleichgewichte, wie Unterschiede in regionalen Durchschnittslöhnen und Arbeitslosigkeitsraten, in der zu erwartenden Weise reagieren. Des weiteren weisen die Resultate einen positiven Zusammenhang des durchschnittlichen Ausbildungsniveaus und der regionalen Arbeitsmobilität auf. Geographische Aspekte, die sich jedoch von Land zu Land unterscheiden können, spielen auch eine wichtige Rolle. Ein generell sehr robustes Ergebnis ist, dass Regionen mit einem sehr hohen Urbanisationsgrad Zentren für regionale Einpendler sind und dementsprechend weniger Auspendler als nicht-urbane Regionen haben. Auch erweisen sich Regionen mit einem hohen Spezialisierungsgrad im Dienstleistungsbereich als Regionen mit hohen Auspendlerströmen. Zudem gibt es Unterschiede zwischen landesinneren Regionen, Grenzregionen, Küstenregionen und Regionen, die an eine EU15 Außengrenze stoßen. Diese Ergebnisse sind weniger robust, insgesamt gesehen jedoch pendeln weniger Arbeitnehmer aus Küstenregionen in die Nachbarschaftsregionen als das der Fall ist für landesinnere Regionen.

Cahier d'études n° 28

The measurement of banking output (and therefore productivity) has long been controversial. This article applies the user cost approach in Fixler and Zieschang (1999) to quarterly reporting data from Luxembourg’s banking sector. This requires associating the flows in the profit-and-loss account to different assets and liabilities in the balance sheet. The user cost of each asset/liability is then calculated as the difference between the rate at which it generates revenues/costs and a “reference rate” representing the opportunity cost of funds. A negative user cost then identifies an asset or liability as an output and a positive user cost identifies it as an input in the production process. In theory, this datadriven approach is capable of combining elements of both the two traditional approaches to measuring banking output (the production and intermediation approaches) since these classify inputs and outputs on an a priori basis. In practice, our results suggest that neither of these conventional approaches is wholly consistent with the data for Luxembourg. We then use multilateral Törnqvist indices to aggregate outputs and inputs separately and show that the resulting series are robust to alternative measures of the reference rate. The difference between the output and input index provides a measure of Total Factor Productivity (TFP) and an implicit price index is also derived. Results suggest that productivity growth in Luxembourg’s banking sector has been high since the mid-1990s, displaying volatile but persistent dynamics and moving pro-cyclically. Productivity varies widely across banks but larger banks (in terms of total assets) tend to be more productive.

JEL classifications: G21, D24, C34

Keywords: banks, total factor productivity, Törnqvist index numbers

Despite strong employment growth, unemployment is rising in Luxembourg. Stronger job competition from non-residents could explain this ‘unemployment paradox’. We construct a small theoretical search unemployment model of the Luxembourg labor market, with job competition between residents and cross-border commuters. We show both analytically and numerically that job competition alone cannot explain the unemployment paradox and we stress the role of the labor market institutions.

Keywords: Job competition, Commuters, Unemployment

JEL classification: J61, J64, R23

Classification du JEL: C32.

Mots clés: GARCH, Corrélations conditionnelles dynamiques.

Cet article a pour objectif d’élaborer, dans une première phase, un indice trimestriel de vulnérabilité financière du secteur bancaire luxembourgeois, permettant ainsi de compléter la batterie d’indicateurs mis en place au sein de la banque centrale pour appréhender la solidité de ce secteur. Dans une seconde étape, le lien entre l’indice construit et l’environnement macroéconomique est exploré à travers une spécification linéaire. Enfin, un modèle est adopté pour la prévision de l’évolution de cet indice. L’approche que nous adoptons pour la construction de cet indice s’inscrit fondamentalement dans la lignée des travaux de Hanschel et Monnin (2005) et Illing et Liu (2006). L’élaboration de cet indice est basée sur un large éventail de variables bilantaires et macrofinancières. Et plusieurs méthodes furent adoptées pour sa construction. Il s’agit de l’indice standard établi avec des pondérations à variance égale, de celui construit selon les percentilles de la fonction de distribution cumulative des variables initiales et enfin, de l’indice élaboré à partirde l’application de la méthode de la composante principale aux données de notre échantillon. Les résultats obtenus révèlent une nette progression de l’indice de vulnérabilité du secteur bancaire luxembourgeois durant la crise financière russe (1998) et au cours de la période 2001-2003. Cette dernière période est caractérisée à la fois par la chute des indices boursiers en 2001-2003 et par l’émergence de la crise financière en Turquie et en Argentine. Quant à la baisse très nette de cet indice durant la période 2003-2006, elle signifie que l’environnement et le comportement du secteur bancaire luxembourgeois en matière de risque furent propices à la stabilité financière. Les résultats prévisionnels obtenus selon notre modèle semblent être en faveur d’une évolution contenue de cet indice. En tenant compte de l’incertitude qui entoure la prévision, la frontière de l’intervalle de confiance reste inférieure aux niveaux historiques les plus élevés observés en 2002 et en 2003. Ceci revient à affirmer qu’en l’absence d’un choc conjoncturel exceptionnel ou d’événements sévères d’ordre systémique, la vulnérabilité du secteur bancaire luxembourgeois demeure faible.

Classification du JEL: G10; E5

Mots-clés: Crise financière; Vulnérabilité financière; Institutions financières; Marchés financiers.

The Luxembourg private sector pension system (“régime général de pension”) is at crossroads. On the one hand, the current budgetary situation of the system appears extremely favourable. On the other hand, projections based on reasonable assumptions suggest that the pension regime is not sustainable over a long-term horizon. Pension benefits are indeed bound to increase steeply when large contingents of crossborder and immigrant employees will retire.

The primary objective of the paper is to assess whether a solution proposed by Modigliani and Muralidhar, where pensioners are gradually transferred from pay-as-you-go (PAYG) to a public fund in accordance with the defined benefit principle, is suitable to the Luxembourg situation. A baseline funding scenario designed in a stepwise manner and under reasonable return assumptions illustrates how fruitful such a solution could be in Luxembourg, provided that a significant prefunding effort takes place at the beginning of the transition period. In the steady state, this scenario would lead to very comfortable reserves and budgetary surpluses with no additional cost in terms of long-term, equilibrium contribution rates. These very favourable results would be achieved in spite of a continuously increasing pension cost ratio induced by ageing and by the gradual retirement of large contingents of cross-border workers. Another particularly attractive feature of the baseline funding scenario presented in the paper – especially in the context of a small and very open economy – is that it would mitigate in an effective way the impact on the pension regime of adverse GDP growth developments. By contrast, PAYG does not provide a sound basis for a social security scheme as contributions are very sensitive to small changes in the key macroeconomic variables. Finally, the baseline funding scenario is reasonably resilient to alternative return or demographic assumptions. However, even the funding system would have to be monitored in a rigourous way. To sum up, the currently favourable situation of the private sector pension regime should be considered as a window of opportunity during which the pension system should set aside the large assets required in order to cover future pension liabilities. This would mark the onset of a virtuous circle, where increasing assets and the related property incomes would offset the rising cost of pension benefits and at the same time mitigate adverse macroeconomic developments.

Keywords: Pensions, funding, defined-benefits, sustainability.

JEL classification: E62, H55, J11.

This paper studies the behaviour of Internet prices. It compares price rigidities on the Internet and in traditional brick-and-mortar stores and provides a cross-country perspective. The data set covers a broad range of items typically sold over the Internet. It includes more than 5 million daily price quotes downloaded from price comparison web sites in France, Germany, Italy, the UK and the US. The following results emerge from our analysis. First, and contrary to the recent findings for common CPI data, Internet prices in the EU countries do not change less often than online prices in the US. Second, prices on the Internet are not necessarily more flexible than prices in traditional brick-and-mortar stores. Third, there is substantial heterogeneity in the frequency of price change across shop types and product categories. Fourth, the average price change on the Internet is relatively large, but smaller than the respective values reported for CPI data. Finally, panel logit estimates suggest that the likelihood of observing a price change is a function of both state- and time-dependent factors.

Keywords: Price stickiness, Internet, price setting behaviour.

JEL: E31, L11.

This paper investigates the issue of the sensitivity of intermediary balances of the profit and loss account of Luxembourg banks to monetary, financial and macro-economic shocks. The analysis is based on a nonbalanced panel with data at quarterly frequency covering the period from 1994 to 2005. The methodology applied is similar to the one of Lehmann and Manz (2005). The estimation results and the simulations seem to indicate that Luxembourg banks are much more sensitive to changes in euro area GDP and to shocks related to the European DJE Stoxx than to monetary shocks. The latter are approximated by a quarterly change of one hundred basis points of the three-month Euribor. The extent of this reactivity should not be overemphasized, as it is not a fundamentally destabilizing factor of the banking sector as a whole.

Keywords: Banks, macroeconomic shocks, Models with panel data

JEL Classification: G21, E32, C33

House prices have grown regularly in Luxembourg since the 1970’s. Since 1997, the average annual growth rate is close to 10%. Since then, concerns about the origin of these developments have repeatedly merged. The aim of this paper is to provide an assessment of the housing market in Luxembourg. More precisely, we wonder whether house price increases are driven by fundamentals or whether a speculative bubble characterizes this market. We resort to a coïntégration analysis where we disentangle residential, non-residential and building plot markets. It is shown that fundamentals seem to have played a role in the growth of house prices. Particularly, construction costs, economic and demographic growth, and monetary policy through interest rates and credits contribute to the developments of house prices. Nevertheless, we should remain cautious regarding our conclusions since the implemented tests are subject to inherent technical pitfalls.

Codes JEL : G12, G21, R21

This paper analyses the pricing behaviour of Luxembourg firms based on survey evidence. Luxembourg firms typically have low market share, many competitors and longstanding customer relationships. Price discrimination is frequently applied. A majority of firms use price review rules that include elements of state dependency. The median firm reviews and changes prices twice a year. The results suggest an almost equal share of firms applying forward- looking, backward-looking and rules of thumb behaviour. The adjustment speed is faster when cost goes up and demand goes down than in the opposite cases. The most relevant theories explaining price rigidity are implicit contracts, cost-based pricing and explicit contracts. Increases in labour and other costs are the most important factors leading to price increases; for price reductions it is price reductions by competitors followed by declining labour costs.

Keywords: Survey data, price setting, price rigidity, adjustment speed JEL

Codes: C21, C22 , C14

Keywords: Fisher hypothesis; Stock market; Markov Switching

JEL classification: G12; E44; E52

This paper uses micro-level price data and analyses the behaviour of consumer prices in Luxembourg. We find that the median duration of consumer prices is roughly 8 months. The median durations of energy and unprocessed food are about 1.5 and 5 months, while prices of services typically change fewer than once a year. For some product types, such as non-energy industrial goods and processed food, a relatively large share of the observed price changes is reverted afterwards. With the exception of services, individual prices do not show signs of downward rigidity. On average, price decreases are as large as price increases. Price changes are determined both by state- and time-dependent factors. Accumulated price and wage inflation, wage adjustment due to indexation, the cash changeover and a larger number of competitors increase the probability of a price change, while pricing at attractive pricing points and price regulation have the opposite effect.

Keywords: Price setting, consumer prices, rigidity, wage indexation, sales

JEL-Codes: E31, C23, C41

This paper analyses Cape Verde's exchange rate policy and investigates whether viable alternatives exist. Cape Verde currently operates a fixed exchange rate regime which, since 1999, links the national currency to the euro. The fixed exchange rate has many benefits, but authorities have to leave interest rates high in order to attract foreign capital, which has inhibited private investment and economic growth; the appreciation of the euro in 2002 and 2003 put the fixed exchange rate under additional strain. This issue is addressed by contemplating whether interest rates can be reduced in the context of the current exchange rate regime, and what costs and benefits are associated with a regime change that enables a reduction in interest rates. The analysis strongly suggests that it is not so much the exchange rate regime that is to blame for high interest rates, but rather a structural problem in the banking sector. Consequently, the policy conclusion reached in this paper is that although changing the current exchange rate policy might reduce interest rates, structural reforms would be more appropriate to tackle the problem at hand.

Keywords: Cape Verde, exchange rates, interest rates, central banks and monetary policy, remittances, modern asset market model

JEL codes: O55, F31, E40, E58

Dans notre contribution, l’estimation du niveau du taux d’intérêt naturel pour la zone euro et pour le Luxembourg est basée sur l’approche semi-structurelle proposée par Laubach et Williams (2003). Cette approche s’appuie sur un petit modèle macroéconomique combinant une équation d’offre agrégée (courbe de Phillips) et une équation de demande agrégée (courbe IS). Le filtre de Kalman sert à estimer les variables inobservables, tels que le taux d’intérêt naturel, l’écart de production, et la croissance potentielle, en tenant compte de l’évolution des variables observées, en l’occurrence la production, l’inflation, et le taux d’intérêt réel. Ainsi, le taux d’intérêt naturel et la croissance potentielle sont estimés de manière simultanée.

Pour la zone euro, les résultants suggèrent une certaine stabilité du taux d’intérêt naturel depuis 1970 et confirment qu’il a baissé au cours de la dernière décennie. Pour le Luxembourg, le taux d’intérêt naturel estimé est beaucoup plus élevé, signe d’une croissance potentielle plus importante. Les résultats laissent présager que la politique monétaire commune a eu un impact expansif sur les périodes récentes, particulièrement au Luxembourg.

This paper analyses the degree of price rigidity and of inflation persistence across different product categories with particular focus on regulated prices and services for the individual EU15 countries, as well as for the EU15 and the euro area aggregates. We show that services and those HICP sub-indices considered being subject to price regulation exhibit larger degrees of nominal price rigidities, with less frequent but larger price index changes as well as stronger asymmetries between price index increases and decreases. With regard to what extent services and regulated prices contribute to the degree of overall inflation persistence, we find that, for most of the EU15 countries as well as for the EU15 and the euro area aggregates, excluding services from the full HICP results in a reduction in the measured degree of inflation persistence; for regulated indices such an effect is also discernible, albeit to a lesser extent.

Keywords: Price rigidity, inflation persistence, regulated prices, services

JEL Codes: E31, C22, C23, C43

Mots clés: compte de transactions courantes, compétitivité-prix, modèle intertemporel de la balance des transactions courantes.

Classification du JEL: E 21, F32, F41, F47

The aim of this paper is to analyse the degree of inflation persistence in Luxembourg using disaggregate price index data from the Harmonised Index of Consumer Prices. The degree of inflation persistence is then compared to estimates for the EU15 and for the euro area as well as for the individual member countries according to a unified approach. In order to assess the robustness of our estimates both a parametric and a non-parametric measure of inflation persistence is used. Overall, our results suggest a relatively low degree of inflation persistence in Luxembourg. For a large number of sub-indices we are not only able to reject the unit root hypothesis, but also we find a low degree of inflation persistence relative to other EU15 countries and relative to the EU15 and euro area aggregates. For Luxembourg as well as the other EU15 countries, our results suggest substantial heterogeneity in the degree of inflation persistence across sub-indices. We find some support for the presence of aggregation effects,    both across indices and countries. Structural break tests for all EU15 countries suggest the presence of structural changes in the inflation process owing to the inception of the single monetary policy and/or to the modified treatment of sales.

Keywords: Inflation persistence, Mean reversion, Aggregation effect, Structural breaks

JEL Codes: E31, C21, C22, C14

This paper explores the compatibility of Eurosystem monetary policy with Taylor-type rules. The Taylor rule aims at giving central bankers guidance in the setting of monetary policy against the background of macroeconomic instability. The initial specification of the Taylor rule determines a level of short-term interest rates considered compatible with the objective of price stability and the elimination of the output gap. The spread between the "Taylor rate" and observed interest rates is an indicator of the appropriateness of monetary policy with respect to macroeconomic fundamentals. The first part of the paper focuses on the advantages and disadvantages of the Taylor rule in its original form. Its application to monetary policy in the euro area since monetary union yields inconclusive results. Modifications to the weights and/or the inclusion of additional variables allow it to track observed interest rates more closely. The second part of the paper deals with differential effects of the single monetary policy. An application of selected Taylor-type rules for the euro area to Luxembourg tends to support the existence of differential effects.

Mots clés: Inflation, Les règles de politique monétaire, Ecart de production

Classification du JEL : E3, E52, E58

This paper uses individual supermarket prices and analyses to what extent absolute deviations from the law of one price are attributable to transaction costs. The results indicate that absolute percentage price differences are increasing in distance, but at a decreasing rate. Similarly, crossing borders increases price deviations, while being inside the former Belgian-Luxembourg monetary association has the opposite effect. This result nurtures the hopes that the euro may be able to reduce regional and cross-border price differences in the long term. Furthermore, larger differences in packaging sizes result in larger price deviations, while the opposite is the case for prices observed within the same retail group.

Keywords: Euro, Price Dispersion, Price Convergence, Law of One Price

JEL Classification: E31, F36, R11

This paper investigates the risks and risk mitigation techniques which are in use within the Luxembourg-based payment and securities settlement systems and evaluates what could be the impact of risk analysis on central bank oversight. It indicates how overseers can make use of different tools to assess the availability and appropriateness of risk mitigation techniques implemented by market practitioners and to further promptly react in case of contingency.

Le Cahier d'études a pour objet de fournir une première évaluation de la situation budgétaire du régime général de pensions au Luxembourg. Une telle analyse est certes confrontée à de multiples difficultés, tant il est difficile de prévoir l'évolution des déterminants de l'équilibre financier du régime à plusieurs décennies de distance. Cependant, l'économie luxembourgeoise présente diverses caractéristiques, dont les effets se déploient sur un horizon de très long terme. Dans un tel contexte, il importe de développer des outils permettant de baliser les évolutions futures, à défaut de les prévoir. La BCL a élaboré un tel outil, dont l'articulation logique et les principaux enseignements sont brièvement décrits ci-dessous.

1.   La situation présente du système de pensions luxembourgeois

La situation budgétaire présente du système de pensions paraît excellente. Les recettes du système ont été systématiquement supérieures aux dépenses au cours de la période 1990-2001, à raison d'environ 2% du PIB. La sédimentation d'excédents sur une longue période a donné lieu à de confortables réserves, qui se sont élevées à 22% du PIB à la fin de l'année 2001.

2.   Diverses caractéristiques propres au Luxembourg

Diverses caractéristiques de l'économie luxembourgeoise sont à la base des excellents résultats budgétaires du régime général. Il s'agit de la forte proportion de travailleurs frontaliers, d'un solde migratoire très positif et enfin de l'importance du secteur financier.

La population des frontaliers est relativement jeune, de sorte qu'elle devrait continuer à alimenter une progression soutenue des cotisations au cours des prochaines années. Cette situation est cependant appelée à se modifier lorsque les importants effectifs de frontaliers parviendront à l'âge de la pension. La même situation prévaut, mutatis mutandis, pour les arrivées de travailleurs étrangers. Une troisième caractéristique du Luxembourg est sa potentielle vulnérabilité aux chocs macroéconomiques ou financiers. Ce facteur a opéré en faveur du régime de pensions au cours des années récentes, en particulier de 1996 à 2000. Une évolution en sens inverse pourrait cependant se manifester dans le futur.

3.   L'outil de projections développé par la BCL

Le simulateur élaboré à la BCL comprend un modèle démographique, qui permet d'inférer l'évolution de la population par sexes et classes d'âge sur la période 2001-2085. Sur la base de taux d'activité, la population active assurée au système de pensions privé, l'évolution de la masse salariale et enfin le montant des cotisations futures peuvent être dégagés. Les dépenses de pensions sont également estimées à partir de la population répartie par sexes et classes d'âge. Sur la base de matrices de statuts, le nombre de pensionnés futurs est inféré. Les dépenses de pensions totales sont obtenues après multiplication par des pensions moyennes représentatives. Le même schéma est appliqué aux travailleurs frontaliers.

4.   Les perspectives d'évolution à long terme du système privé de pensions

4.1.   La projection de référence

Les hypothèses qui président à la simulation de référence sont décrites en détail dans le Cahier d'études. Il est notamment supposé que les arrivées nettes de frontaliers et d'immigrants demeureraient à des niveaux relativement élevés, soit respectivement 7000 et 4000 personnes par an à partir de 2005. Par ailleurs, la projection de référence inclut les mesures adoptées dans la foulée du Rentendësch et suppose le maintien du taux de cotisation à 24% des revenus cotisables.

La soutenabilité à terme du système de pensions paraît précaire sous ces hypothèses. A partir de 2005, la reprise des entrées de frontaliers et l'immigration soutenue conforteraient certes la croissance économique, ce qui tendrait à réduire l'importance relative des dépenses tout en favorisant les recettes. Il en résulterait une augmentation des excédents et une hausse soutenue du niveau des réserves, qui culmineraient à près de 50% du PIB en 2028. Les excédents primaires commenceraient cependant à décliner dès 2015, tandis que la capacité de financement diminuerait à partir de 2021. Elle cèderait la place à un besoin de financement dès 2041, qui s'élèverait à quelque 8% du PIB à l'issue de la période de projection. Dans un tel contexte, un premier endettement ferait son apparition en 2055. Il atteindrait 60% du PIB vers 2070 et 109% en 2085.

4.2.   Première variante: plafonnement de la population à 700 000 habitants

La population obtenue dans le cadre de la projection de référence dépasserait le seuil des 700 000 habitants à partir de 2048. Une projection a été effectuée, qui consiste à plafonner la population à 700 000 habitants en annulant les soldes migratoires à partir de 2049. Un tel plafonnement pourrait par exemple résulter d'une saturation des infrastructures publiques, ou encore d'une capacité de logement ou de transport insuffisante. Ce plafonnement affecterait très négativement l'équilibre du système de pensions. La dette et le besoin de financement atteindraient en effet 170% et 13,5% du PIB, respectivement, en 2085.

4.3.   Deuxième variante : ralentissement des arrivées de frontaliers

Les résultats de la simulation de référence dépendent de façon cruciale de l'hypothèse d'un maintien à un niveau élevé des arrivées nettes de frontaliers. Un scénario alternatif consiste à abaisser ces arrivées nettes à 4000 individus par an à partir de 2005. Dans ce contexte, le premier endettement se manifesterait dès 2045 et la dette du régime privé de pensions se monterait à 132% du PIB en 2085. Cet impact négatif s'éroderait cependant fortement à partir de 2057 du fait de la diminution du nombre de frontaliers pensionnés.

4.4.   Evaluation du coût du Rentendësch

Le coût statique en année pleine des mesures de la loi du 28 juin 2002 peut être estimé à 0,6% du PIB. Cependant, une telle estimation statique ignore les interactions entre les nouvelles mesures et l'environnement macroéconomique et démographique, de même que leur impact sur les charges d'intérêt. Compte tenu de ces effets indirects, le coût des mesures atteindrait 1% du PIB dès 2022 et près de 3% en 2085, ce gonflement étant largement imputable aux charges d'intérêt additionnelles. En l'absence des nouvelles mesures, la dette projetée ne se serait établie qu'à 61% du PIB en 2085, au lieu de 109% dans la projection de référence.

4.5.   Analyses de sensibilité des résultats

Diverses projections alternatives permettent d'évaluer la stabilité des résultats et de mieux illustrer la façon dont opèrent divers déterminants de l'équilibre du système privé de pensions. Les résultats de ces projections, qui font l'objet de commentaires détaillés dans le Cahier d'études, mettent en évidence l'effet durable d'une accélération même temporaire d'une hausse des salaires, l'importance cruciale d'une amélioration du taux de rendement des réserves, la sensibilité de la situation budgétaire au taux de participation des femmes au marché du travail et, enfin, les retombées très favorable d'un accroissement du taux de fertilité.

Conclusion

Les résultats brièvement décrits ci-dessus soulignent l'importance d'une surveillance proactive de l'évolution de la situation budgétaire du système de pensions. Il serait hasardeux de n'adopter d'éventuelles mesures de correction que lorsque les réserves de pensions auront significativement décliné. Il serait alors trop tard pour contrer avec succès les facteurs de dérive des coûts, ce qui contraindrait les autorités à hausser significativement les cotisations sociales.

Ces résultats mettent également en évidence la nécessité d'une gestion rigoureuse du budget de l'Etat central. Les excédents de la sécurité sociale sont vraisemblablement appelés à s'étioler à moyen terme. Dans un tel contexte, un déficit structurel de l'Etat central contrarierait davantage qu'actuellement le respect des exigences européennes, qui se rapportent aux administrations publiques considérées dans leur globalité (Etat central, communes, sécurité sociale).

This paper investigates the transmission of monetary policy in Luxembourg. It is the first empirical analysis conducted for Luxembourg firm-level data. The results indicate that the sales accelerator may be at work. A very robust result is the negative effect of the user cost of capital on firms' investment ratio. Changes in user cost are significantly affected by changes in the monetary policy indicator. In addition, firm specific balance sheet characteristics, such as the lagged cash stock to capital ratio influence the investment behaviour according to the broad credit channel theory. It is shown that young firms, in particular, are more sensitive to user cost changes, sales growth and the lagged cash to capital ratio.

JEL Codes: D21, D92, E22, E52

Keywords: Investment, User Cost of Capital, Credit Channel, Panel Data

The output gap is defined as the difference between the observed level of an economy’s output and its trend or potential level. In the short term, an economy can produce above its potential level (a positive output gap) through unusually high levels of labour force participation, capacity utilisation, or technical progress. However, a positive output gap tends to generate inflationary pressures on the markets for factors of production. Once inflation accelerates, output will have to fall below its potential level (a negative output gap) to increase available resources and reduce the pressure on prices. Therefore, measures of the output gap are often used in macroeconomic analysis to assess current and future levels of inflationary pressures in the economy. This study reviews several of the many alternative methods of estimating output gaps and applies six of these to annual data for Luxembourg. These different measures of the output gap are then compared and evaluated in terms of their contribution to inflaton forecasting. Methods based on unobserved components models tend to do better than simpler, better known methods (i.e. linear trends, the Hodrick-Prescott filter). Multivariate methods that consider the simultaneous evolution of several different economic variables tend to do better than univariate methods that limit themselves to the output series itself.

This study analyses productive efficiency in banking and scale and scope economies using stochastic frontier analysis. The European banking sector underwent great changes during the period under study (1995-2000), featuring a marked increase in competition. Empirical results suggest that Luxembourg-based banks in our sample are relatively efficient and that their good performance can in part be explained by independent variables in the cost function and by technological progress. Estimated scale elasticities do not provide evidence in favour of economies of scale. However, test results generally cannot reject the existence of cost-complementarities across the following output pairs: deposits-loans, loans-securities, and deposits-securities.

Mots Clés: X-efficiency; Stochastic Cost frontier; Banking; Panel data.

Classification JEL: G21, D21, C23.

Though stock prices are commonly not considered an integral part of central banks’ monetary policy strategy, financial asset prices are highly relevant because they exert important impacts on inflation, on the real sphere of the economy, and on the financial system. This paper illustrates the evolution of selected primary and secondary equity markets and elaborates divergences and similarities between the pricing of old and new economy stocks. It is shown that the valuation of new economy stocks is subject to enhanced contingency. Prices of new economy stocks ceteris paribus react more sensitively to new information and modifications to external assumptions. From both a microeconomic as well as a macroeconomic point of view, the growth projections implicit in price earnings ratios observed in recent years seem unrealistic. Furthermore, from a utility maximising perspective, it seems unlikely that the observed shift in investment away from old economy stocks and into new economy stocks could have been achievable without a change in the aggregate risk preference. Panel regression analysis based on 219 EURO STOXX firms, though, confirms a significant impact of firm-specific and macroeconomic fundamentals on monthly returns for old economy companies as well as for telecommunication, media and technology (TMT) firms. The null-hypothesis of no statistically significant difference between TMT firms and non-TMT firms with respect to the role of firm-specific and macroeconomic fundamentals in explaining monthly stock returns is rejected. While theoretical considerations and empirical findings suggest that the monetary policy stance remains an important factor driving equity valuation, the growing passion for stocks and the more volatile pricing of new economy stocks bear important implications for central bank policy making.