The impact of the cost-of-living crisis on European households

Chafwehé, Boris; Ricci, Mattia; Stöhlker, Daniel Working Paper The impact of the cost-of-living crisis on European households JRC Working Papers on Taxation and Structural Reforms, No. 01/2024 Provided in Cooperation with: Joint Research Centre (JRC), European Commission Suggested Citation: Chafwehé, Boris; Ricci, Mattia; Stöhlker, Daniel (2024) : The impact of the cost-ofliving crisis on European households, JRC Working Papers on Taxation and Structural Reforms, No. 01/2024, European Commission, Joint Research Centre (JRC), Seville This Version is available at: https://hdl.handle.net/10419/299565 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ The Impact of the Cost-of-Living Crisis on European Households JRC Working Papers on Taxation and Structural Reforms No 01/2024 Chafwehé, B. Ricci, M. Stöhlker, D. 2024 This publication is a working paper to provide evidence-based scientific support to the European policymaking process. The contents of this publication do not necessarily reflect the position or opinion of the European Commission. Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use that might be made of this publication. For information on the methodology and quality underlying the data used in this publication for which the source is neither Eurostat nor other Commission services, users should contact the referenced source. The designations employed and the presentation of material on the maps do not imply the expression of any opinion whatsoever on the part of the European Union concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. EU Science Hub https://joint-research-centre.ec.europa.eu JRC136870 Seville: European Commission, 2024 © European Union, 2024 The reuse policy of the European Commission documents is implemented by the Commission Decision 2011/833/EU of 12 December 2011 on the reuse of Commission documents (OJ L 330, 14.12.2011, p. 39). Unless otherwise noted, the reuse of this document is authorised under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence (https://creativecommons.org/licenses/by/4.0/). This means that reuse is allowed provided appropriate credit is given and any changes are indicated. For any use or reproduction of photos or other material that is not owned by the European Union permission must be sought directly from the copyright holders. How to cite this report: Chafwehé, B., Ricci, M. and Stöhlker, D., The Impact of the Cost-of-Living Crisis on European Households, JRC Working Papers on Taxation and Structural Reforms No 01/2024, European Commission, Seville, Spain, 2024, JRC136870. This paper was updated in May 2024. 1 Abstract We study the impact of the recent cost-of-living crisis on European households using data on individual consumption, income, and wealth. We account for the various channels through which inflation affects individual households and for the monetary and fiscal policy responses to the inflationary shock. Our results indicate that on average pension-age households lost nearly three times as much as their working-age counterpart, due to the devaluation of their nominal wealth. Along the income distribution, differences in nominal asset holdings and in the evolution of nominal incomes imply that the inflationary shock was regressive for working-age households and mostly flat for pension-age households. Overall, high-income working-age households with mortgage debt gained the most from the inflationary surge, while older individuals with large nominal asset positions were those for which the largest losses were recorded. Fiscal policy measures were able to partially offset the impact of the crisis on the most vulnerable households. The interest rate response to the crisis partially offset the losses recorded by households with large nominal asset positions. 2 Acknowledgements We wish to thank Maximilian Freier from the European Central Bank for a thorough review of this work. The authors are indebted to Salvador Barrios for his advice and guidance. Furthermore, the authors gratefully acknowledge helpful comments from Gonzalo Paz-Pardo, Matteo Salto, Boromeus Wanengkirtyo, and seminar participants at the Fiscal Policy Modelling Workshop in Seville. All errors are our own. Authors Boris Chafwehé Mattia Ricci Daniel Stöhlker 3 Executive Summary Policy context The paper discusses the impact of the recent inflation surge on Eurozone households, a phenomenon driven by the COVID-19 s supply disruptions and the Russian invasion of Ukraine. The increased cost of living, particularly due to spikes in energy prices, has caused one of the most severe financial strain on households in the developed world in decades, with varying effects on individuals depending on the composition of their consumption baskets, income, and wealth. Governments in the Eurozone have responded to the crisis with fiscal measures estimated to cost around 2% of GDP. Some of these measures were particularly designed to shield vulnerable groups from the impact of inflation. Meanwhile, the European Central Bank has raised interest rates to historical highs, affecting both borrowers and lenders. Those policy responses constituted an integral part of the cost-of-living crisis period, as they explicitly tackled the inflationary shock and were key determina wealth. Main analysis In this paper, we study the impact of the cost-of-living crisis on Eurozone households, considering both the effects of the inflationary shock and those of the fiscal and monetary policy responses. Inflationary shocks have an immediate effect on households through three main channels: the Fisher channel (influencing net creditors and debtors differently due to nominal contracts), the relative consumption channel (due to heterogeneous consumption patterns affecting individual exposures to inflation), and the nominal income channel (due to the devaluation of sticky nominal incomes). By utilizing various data sources, including the Household Finance and Consumption Survey (HFCS) and the EUROMOD microsimulation model, the paper quantifies these effects across the income distribution, the demographic status and other population groups. Our analysis crucially incorporates the effects of monetary and fiscal policy responses. The monetary policy Unhedged Interest Rate Exposure (URE), which measures financial gains or losses following interest rate changes depending net financing needs, which are influenced by their portfolio compositions. Fiscal policy effects are assessed using microsimulation techniques to estimate the cushioning effects of government measures, including both price and income-side interventions. The paper positions itself within the literature that examines the heterogeneous effects of the recent cost-ofliving crisis on European households. While related studies have explored various aspects in isolation, this paper provides a comprehensive assessment by considering the direct effects of inflation together with effects arising from fiscal and monetary policy responses. It extends previous research by providing a comprehensive cross-country analysis, incorporating the the impact of interest rate increases on wealth, and by highlighting the importance of characteristics like home ownership in driving the heterogeneous effects of the crisis on European households. Key conclusions We find that pension-age households lost nearly three times as much as their working-age counterpart due to the devaluation of the nominal wealth they accumulated during their life cycle. Differences in nominal balances and in the evolution of nominal incomes from different sources further imply that the inflationary shock was regressive among working-age households but mostly flat among the pension-age. In most cases, the impact of inflation through the Fisher and nominal income channels was an order of magnitude larger than the relative consumption channel, which has been the focus of much of the related literature. Looking at the impact of the fiscal and monetary response, interest rate increases partially offset the losses made by households with large nominal asset positions, mostly pension-age households. In contrast, the extraordinary fiscal measures adopted in response to the crisis were able to partly offset the negative impact of inflation on the lowest-income households. Nonetheless, in several countries, large losses remain among the poorest households and those of pension-age. Finally, we extend our results to consider the role of wealth composition in shaping the impact of the cost-ofliving crisis on population sub-groups. We find that households with fixed-rate mortgages and wealthy handto-mouth households are the biggest winners of the cost-of-living crisis, whereas pension-age households are the main losers. The Impact of the Cost-of-Living Crisis on European Households* Boris Chafweh´ e†Mattia Ricci‡Daniel St¨ ohlker§ May 15, 2024 Abstract We study the impact of the recent cost-of-living crisis on European households using data on individual consumption, income, and wealth. We account for the various channels through which inflation affects individual households and for the monetary and fiscal policy responses to the inflationary shock. Our results indicate that on average pension-age households lost nearly three times as much as their working-age counterpart, due to the devaluation of their nominal wealth. Along the income distribution, differences in nominal asset holdings and in the evolution of nominal incomes imply that the inflationary shock was regressive for working-age households and mostly flat for pension-age households. Overall, high-income working-age households with mortgage debt gained the most from the inflationary surge, while older individuals with large nominal asset positions were those for which the largest losses were recorded. Fiscal policy measures were able to partially offset the impact of the crisis on the most vulnerable households. The interest rate response to the crisis partially offset the losses recorded by households with large nominal asset positions. Keywords: Inflation Heterogeneity, Monetary and Fiscal Policy, Euro Area. JEL classification: G51, D31, E31. *The views expressed in this paper are those of the authors and should not be attributed to the European Commission, the Bank of England or its committees. The authors gratefully acknowledge helpful comments from Salvador Barrios, Maximilian Freier, Gonzalo Paz-Pardo, Matteo Salto, Boromeus Wanengkirtyo, and seminar participants at the Fiscal Policy Modelling Workshop in Seville. †Bank of England. E-mail: [email protected]. ‡European Commission (JRC Seville). E-mail: [email protected]opa.eu. §European Commission (JRC Seville). E-mail: [email protected]opa.eu. 4 1 Introduction The recent surge in inflation – the result of post-pandemic supply disruptions and the Russian invasion of Ukraine – has had a profound impact on household finances across many regions of the globe. Households in the Eurozone were particularly affected by the shocks to energy supplies and the ensuing price increases. Inflation has eroded the real value of nominal incomes and wealth, which has challenged the ability of households to pay for consumption, thereby generating one of the most severe cost-of-living crisis since decades. Crucially, the crisis has affected households in a heterogeneous way. In particular, differences in consumption patterns, sources of income, and the level and composition of wealth brought about substantial differences in the way individual households were impacted by inflation. The inflationary shock triggered a bold policy response. Governments across the Eurozone adopted measures to protect households against the effects of inflation, especially the most vulnerable population groups. These fiscal measures are estimated to have cost some 2% of GDP in years 2022 and 2023 (Ba´ nkowski et al.,2023). On the monetary policy side, the European Central Bank raised interest rates to unprecedented levels, increasing the financing cost of loans and mortgages but also the rate of returns for households re-investing their savings. Those policy responses constituted an integral part of the cost-of-living crisis period, as they explicitly tackled the inflationary shock and were key determinants of its impact on households’ wealth. Accounting for them should therefore be part of any assessment of this crisis. In this paper, we study the impact of the cost-of-living crisis on Eurozone households, considering both the effects of the inflationary shock and those of the fiscal and monetary policy responses. As discussed in Cardoso et al. (2022), inflationary shocks have an immediate effect on households through three main channels: (i) the Fisher channel, due to the fact that some households are net creditors and others are net debtors in contracts denominated in nominal terms; (ii) the relative consumption channel due to differences in consumption patterns across households, which gives rise to differences in effective individual inflation rates; and (iii) the nominal income channel, that accounts for the devaluation of nominal incomes in the presence of nominal rigidities. We quantify the effects of those channels on households in the Eurozone across the income distribution, using data from the Household Finance and Consumption Survey (HFCS), combined with the Household Budget Survey (HBS), EUROMOD (the micro-simulation model of the European Union) and its underlying EU-SILC data. 5 We consider in addition the impact deriving from the monetary and fiscal policy responses to the shock. On the monetary policy side, we know from Auclert (2019) that interest rate increases impact households’ balance sheets through the so-called ‘Unhedged Interest Rate Exposure’ (URE). The URE provides a measure of the financial gain/loss that households suffer following an increase in the interest rate, depending on their net financing needs. These losses depend on the composition of households’ portfolios, and in particular the maturity of their assets and liabilities. We construct the URE at the household level using HFCS data to quantify the impact of the interest rate response on households across population subgroups. On the fiscal policy side, we draw from the recent work of Amores et al. (2023a) to quantify the cushioning effects of fiscal measures on the “income-side”. To quantify measures on the “priceside”, instead, we exploit differences between standard inflation figures and those calculated at constant taxes.1 We find that pension-age households lost nearly three times as much as their working-age counterpart due to the devaluation of the nominal wealth they accumulated during their life cycle. Differences in nominal balances and in the evolution of nominal incomes from different sources further imply that the inflationary shock was regressive among working-age households (affecting low-income households the most) but mostly flat among the pension-age. In most cases, the impact of inflation through the Fisher and nominal income channels was an order of magnitude larger than the relative consumption channel, which has been the focus of much of the related literature cited below. Looking at the impact of the fiscal and monetary response, interest rate increases partially offset the losses made by households with large nominal asset positions, mostly pension-age households. By contrast, the extraordinary fiscal measures adopted in response to the crisis were able to partly offset the negative impact of inflation on the poorest households. Nonetheless, in several countries, large losses remain among the poorest households and those of pension age. Finally, we extend our results to consider the role of wealth composition in shaping the impact of the cost-of-living crisis on population sub-groups. In particular, we look at the homeownership, mortgage and hand-to-mouth status of households. We find that accounting for the home-ownership and mortgage status helps explain most of the differences in the effects of inflation on household wealth across age groups. We show that there is substantial heterogene1Throughout the paper, we follow the terminology used by Amores et al. (2023a) and distinguish between fiscal policy interventions aimed at directly mitigating the effective prices paid by households - the price-side measuresand interventions aimed at supporting household incomes, the income-side measures. 6 taxes and fiscal transfers. Formally, we show in the Appendix that the fiscal impact can be expressed as: da(τ) j,t=d˜ Tj,t |{z} Income-side measures −"cj,t dτc j dτc−1!+NNPj,t+ (1−λj,t)y(t) j,t−1#dτc t | {z } Price-side measures: spending & wealth/income effects (8) The first element, d˜ Tj,t, captures income-side measures adopted by governments to support household income directly. The exact design of those measures and their size varied across countries, but in most cases they were targeted towards low-income households. The second term of Equ. (8) relates to the effects of price-side measures, with which governments tried to dampen price pressures through changes in indirect taxes, subsidies, discounts, etc. In our framework, those measures – modeled through a change in consumption tax rates dτc– have a direct effect on consumption prices. Therefore, they affect the effective inflation rates faced by households, mitigating the impact of inflation through the relative consumption, Fisher and nominal income channels described above. 2.5 The Monetary Policy Impact Interest rate fluctuations have a direct effect on the interest income flows received or paid by households. Our analysis focuses exclusively on such direct (first-order) interest rate effects and disregards the effect that monetary policy has on economic activity and inflation. As described in Auclert (2019), the impact of interest rate changes on households’ balance sheets can be summarised through the so-called ‘Unhedged Interest Rate Exposure’ (URE). The URE is defined as the difference between maturing assets and liabilities at a given point in time. Maturing assets include households’ net income, and maturing liabilities include households’ current consumption. In net terms, it is the resource flow available to households to be saved or the amount required to be borrowed by households, over an interval of time, that is exposed to current changes in interest rates. Obviously, it is important in this context to consider each asset’s and liability’s maturity, since longer maturities partially protect households against transitory interest rate changes, as in the case of mortgage contracts with fixed interest payments. Such assets and liabilities are considered to be ‘hedged’ against a change in the interest rate, as compared to ‘unhedged’ ones with short maturities. Assuming a complete pass-through of the policy rate into retail rates for deposits and loans and bond prices, the in13 dividual interest rate exposure translates one-to-one into a (direct) effect on individual wealth, following a change in the policy rate. More formally, we show in Appendix (A) that the impact of changing interest rates can be summarised as follows: da(R) j,t=UREj,tdR (9) where UREj,t=B(t) t−1 Pt+b(t) t−1+yj,t−Tj,t−∑k Pk,t Ptcj,k,tis the difference between the maturing assets and maturing liabilities of the household. Households with a positive URE, e.g. those who hold large amounts of sight account deposits or other short-term instruments, benefit from a rise in interest rates. By contrast, households with a negative URE, e.g. those holding large amounts of adjustable-rate mortgages, lose from an increase in interest rates through higher interest payments on their maturing debt position. 3 Empirical Strategy We analyse the impact of the cost-of-living crisis on households in six Eurozone countries: Germany, France, Italy, Spain, Portugal, and Greece.7For this purpose, we combine data from different sources to quantify the effects of inflation, fiscal policy and monetary policy derived in the previous section. Data on household (gross) income and consumption, wealth and its composition are obtained from the third wave of the Household Finance and Consumption Survey (HFCS), containing data for the year 2017.8Information on the composition of households’ consumption basket are obtained from the 2015 Household Budget Survey (HBS). Finally, the EUROMOD micro-simulation model, together with its Indirect Tax Tool (ITT) extension,9are used to: (i) translate gross incomes from the HFCS into disposable incomes, (ii) construct a measure of nominal income growth during our period of analysis, and (iii) simulate the effects of the fiscal measures. 7These countries together represent some 85% of the Eurozone GDP; we then use them as a proxy for the Eurozone as a whole. 8We deliberately refrain from using data from the most recent fourth wave of the HFCS survey, which was conducted between the first half of 2020 and the first half of 2022. Given the disruptive nature of the COVID-19 pandemic and its impact on household balance sheets (e.g. through income losses and (in)voluntary savings), we did not consider this data to be the most reliable for the exercise we conduct in this paper. 9EUROMOD and ITT, in turn, make use of the EU-SILC and HBS as underlying data sources. 14 Table (1) provides a summary of the various data sources we use to compute the effects of the cost-of-living crisis across population groups. In what follows, we describe our empirical strategy in more detail. The inflationary shock We consider the period 2021M6-2023M12 as the main period of analysis (i.e. period tin the language of the framework presented above), as Eurozone inflation started to surge in the second half of 2021, and – while not yet fully back to the 2% target – had already gone down significantly by the end of 2023. To construct the ‘surprise inflation‘ measure, we make use of the 2021Q3 wave of the Survey of Professional Forecasters (which reflects inflation expectations of the financial sector as of mid-2021) to subtract the expected cumulative inflation from the realised inflation in our period of analysis.10 To compute the direct effects of inflation, we further use the HICP at constant tax rates (HICPCT), which is a variant of the HICP that excludes the impact of changes in consumption taxes, such as value-added tax (VAT) and excise duties, on consumer prices. We use this measure to separate the direct effects of inflation from those arising from the fiscal response appearing in Equ. (8), which are reported separately.11 Population groups Our main results (presented below in Section (4)) are provided for population subgroups that vary along the income and age dimensions. First, we group households into deciles of gross income, separately for each country. To obtain a measure of household disposable income from the HFCS data (which only contains information on market incomes), we make use of EUROMOD to calculate the ratio between gross and disposable income by decile of market income (EUROMOD,2023). Once households have been assigned to an income decile group, we assign them to an age group based on the reported age of the household head. We consider two main age groups: working-age households, with age below 65, and retirement-age households, with age above 65. We then end up with 20 groups in total for each country. In Table (2), we report the joint distribution of households into those twenty groups in row (A), pooling all countries together. 10The implied cumulative inflation forecast for our period of analysis was of 2.494% in the 2021Q3 SPF. In Appendix (D.1) we provide a figure comparing the inflation path implied by this forecast to realised inflation. The gap between the two is equal to 12.7 percentage points. 11One limitation of using the constant-tax inflation measure is that it does not include price-side measures which are not tax-related, such as price-caps and reimbursement for higher energy prices. This means that the fiscal effects that we report in the paper are effectively a lower bound on the actual effects of the fiscal response to inflation. Headline and constant-tax cumulative inflation rates during our period of interest are summarised in Table (1). 15 We observe from the table that, on average, there is a higher share of households with young individuals in high-income groups, reflecting the fact that working-age individuals, who obtain a large fraction of their income as labour income, earn more than retirees. The equivalent figures for individual countries are available in the Appendix C.1 of this paper. In Section (5), we extend our results by studying the effects of the crisis on various population subgroups according to the composition and the level of their wealth. Rows (B) to (E) of Table (2) show the fraction of households according to wealth characteristics such as the homeownership and mortgage status, by income and age group. We observe that households with young individuals are on average less likely to be home-owners (especially at the lower end of the income distribution), more likely to have a mortgage, and more likely to be considered as ‘hand-to-mouth’ (i.e. having insufficient holdings of liquid assets to smooth out consumption in the event of averse income shocks).12 Consumption basket composition To compute the effects of inflation arising from the relative consumption channel, we need to account for heterogeneity in consumption baskets to compute effective inflation rates for our age/income groups of interest. To do so, we use the Household Budget Survey (HBS), which provides information on individual consumption expenditures by COICOP consumption good category.13 HBS data on net incomes are used to group households into income deciles. The HBS also contains data on age, so we are able to compute age and decile-specific inflation rates by country, that we can then plug into Equ. (7) when assessing the effects of inflation. Nominal income growth To obtain the effects of inflation through the nominal income channel, we need to compute the value of λjfor each of our groups of interest. As reflected in Equ. (2), λjdenotes the fraction of household income that grows with inflation. To compute it from the data, we assume that all the growth in nominal market incomes during our period of interest was due to inflation. We then compute growth in disposable incomes from EUROMOD, using the following strategy. We first use EUROMOD ‘uprating factors’ to adjust nominal incomes in the latest EU-SILC data (dating to 2021) to approximate their values in 2022 and in 2023.14 Then, to obtain disposable income growth abstracting from the policy changes 12The Appendix provides more details on how individuals are assigned to the various subgroups we consider. 13Data on inflation rates at the country level and the COICOP 4-level at constant tax rates are obtained from the ECB’s Harmonised Index of Consumer Prices (HICP). 14EUROMOD ‘uprating factors’ are mostly based on Eurostat data on nominal income growth by sector of activity and income source. The full list of uprating factors used for each country and the under16 during our period of analysis, we compute it as: 1+∆Y=EM2023(y2023) EM2023(y2022)×EM2022(y2022) EM2022(y2021m6)(10) where EMt(·)summarises the EUROMOD tax-benefit calculator using policy rules of year t to translate market incomes yinto disposable incomes.15 For each year t, market incomes yare updated, using uprating factors between 2021M6 and year t. The obtained values of uprated incomes are denoted ytin the above formula. Note that, by calculating in each year the ratio between disposable incomes from uprated and not uprated market incomes, we are effectively eliminating the effect of policy changes. Finally, we compute the value of λfor each country/population group as λj=∆Yj π, using ∆Yjcomputed from (10), and where πis the country’s aggregate cumulative inflation rate at constant tax rates for the period 2021M6 to 2023M12. Net nominal asset positions and Interest rate exposures The net nominal asset positions (NNP) and the ‘Unhedged Interest Rate Exposure’ (URE) of individuals, which are necessary to estimate the effects of inflation and monetary policy through the Fisher and interest rate channels (see Equ. (7) and (9)), are computed at the household level from the HFCS data, before aggregating them by population subgroups. Details on how those variables are computed are provided in Appendix (B.1) and (B.2). Fiscal support As mentioned in Section (2), to assess the effects of the fiscal response to the crisis, we distinguish between the ‘price-side’ (interventions aiming at reducing the prices paid by consumers), and ‘income-side’ (fiscal transfers aiming at supporting household incomes). To compute the effects of price-side measures, we make use of the regular HICP inflation series together with the HICP-CT measure (which removes the effect of indirect tax changes, and which we use to compute the direct effects of inflation, as described above) to obtain an implicit measure of the effect of tax changes on inflation, from Equ. (6). We then use the values of τc(using aggregate inflation) and τc j(using group-specific inflation rates) obtained from this procedure to compute the fiscal effects outlined in Equ. (8). lying data sources are documented in EUROMOD country reports, available online at: https://euromodweb.jrc.ec.europa.eu/resources/country-reports. 15Note that the policy rules used to update nominal incomes do not include the effects of the policy response to the cost-of-living crisis, which are reported in the fiscal policy effects d˜ T. 17 To assess the effects of income-side fiscal measures, i.e. the various social benefits and income support measures taken by governments to help households cope with rising living costs – denoted as d˜ Tj,tin Equ. (8) – we draw from the recent work of Amores et al. (2023a), who use microsimulation techniques to estimate the cushioning effect of those measures by income decile for the same subset of countries as the one we consider in this paper. As those values are not available for different age groups, we assign the same effects to working-age and retirementage individuals that are part of the same income decile. Moreover, given that Amores et al. (2023a) provided these calculations only for 2022, we project them to 2023 based on their relative budgetary cost at the macro level.16 under the assumption that the degree of targeting (wrt 2022 measures) did not change. 16The budgetary impact of fiscal measures adopted by governments in 2022 and in 2023 in support of households are drawn from the European Commission’s calculations, summarised in Bethuyne et al. (2022). 18 TABLE 1: Main Data Sources and Inflation Numbers Variable Source/Value Individual exposures Net nominal position (NNP) HFCS Gross Income (Y) HFCS Consumption level (C) HBS & HFCS Gross to disposable income EUROMOD Interest rate exposure (URE) HFCS Inflation effect Nominal income indexation (λj) EUROMOD Aggregate inflation (π) ECB (HICP) Expected inflation (Et−1πt) SPF (2.50%) Effective inflation rate (πj) HBS & COICOP4 π(ECB). Policy response Fiscal response (d˜ Ty) EUROMOD & Amores et al. (2023a). Interest rate response (R) ECB Country HICP inflation HICP-CT inflation France 12.94% 13.48% Germany 16.04% 16.29% Greece 14.49% 14.55% Italy 15.68% 16.19% Portugal 13.01 14.20% Spain 12.66% 14.21% Euro Area 15.18% 15.59% Notes: HFCS: Household Finance and Consumption Survey, 2017 (Wave 3). SPF: Survey of Professional Forecasters, conducted by the ECB (2021Q3). HBS: Household Budget Survey, 2015 wave. EUROMOD is the micro-simulation model for tax-benefit system for the EU27, which uses the EU Statistics on Income and Living Conditions (EU-SILC) as its main data input source. The EUROMOD ITT (Indirect Tax Tool) extension makes use of the Household Budget Survey (HBS) as additional data source. 19 TABLE 2: Population Distribution Across Income Deciles Decile 1 2 3 4 5 6 7 8 9 10 A All WA 7.16% 4.90% 5.77% 6.26% 6.65% 6.79% 7.41% 7.84% 8.10% 8.23% RA 2.86% 5.11% 4.27% 3.74% 3.36% 3.17% 2.59% 2.19% 1.87% 1.76% B Home-owner WA 1.96% 1.50% 2.41% 2.84% 3.47% 4.04% 5.03% 6.02% 6.73% 7.40% RA 1.50% 2.77% 2.74% 2.75% 2.52% 2.61% 2.23% 2.00% 1.69% 1.66% Non owner WA 5.19% 3.39% 3.36% 3.42 % 3.18% 2.76% 2.39% 1.83% 1.37% 0.84% RA 1.34% 2.33% 1.53% 0.99% 0.83% 0.56% 0.36% 0.20% 0.18% 0.09% C Mortgage WA 0.47% 0.41% 0.89% 1.04% 1.53% 1.89% 2.55% 3.05% 3.81% 4.18% RA 0.06% 0.10% 0.12% 0.17% 0.18% 0.17% 0.17% 0.21% 0.26% 0.29% No mortgage WA 1.49% 1.09% 1.51% 1.80% 1.94% 2.15% 2.48% 2.97% 2.92% 3.22% RA 1.44% 2.68% 2.62% 2.58% 2.35% 2.44% 2.06% 1.79% 1.43% 1.37% Non owner WA 5.19% 3.39% 3.36% 3.42 % 3.18% 2.76% 2.39% 1.83% 1.37% 0.84% RA 1.34% 2.33% 1.53% 0.99% 0.83% 0.56% 0.36% 0.20% 0.18% 0.09% D AR mortgage WA 0.22% 0.18% 0.43% 0.44% 0.64% 0.70% 0.78% 1.04% 1.16% 1.17% RA 0.02% 0.02% 0.02% 0.04% 0.04% 0.02% 0.04% 0.04% 0.04% 0.03% FR mortgage WA 0.25% 0.23% 0.46% 0.60% 0.89% 1.18% 1.77% 2.01% 2.65% 3.01% RA 0.05% 0.08% 0.09% 0.13% 0.14% 0.16% 0.13% 0.17% 0.22% 0.26% E Poor HTM WA 2.08% 1.37% 1.09% 0.96% 0.61% 0.50% 0.47% 0.25% 0.13% 0.06% RA 0.42% 0.79% 0.44% 0.21% 0.15% 0.11% 0.05% 0.03% 0.04% 0.00% Wealthy HTM WA 1.12% 0.89% 1.32% 1.46% 1.98% 1.96% 2.30% 2.46% 2.49% 2.40% RA 0.47% 0.57% 0.51% 0.39% 0.37% 0.32% 0.22% 0.20% 0.19% 0.15% Non HTM WA 3.95% 2.63% 3.35% 3.84% 4.06% 4.33% 4.64% 5.14% 5.49% 5.78% RA 1.96% 3.75% 3.32% 3.14% 2.83% 2.74% 2.32% 1.96% 1.63% 1.60% F Effective πWA 17.35% 16.57% 16.27% 16.19% 15.94% 15.76% 15.71% 15.48% 15.27% 14.91% RA 14.84% 15.19% 14.88% 15.04% 14.48% 14.59% 14.34% 14.27% 14.37% 13.82% Notes: WA: working-age (<65 years old) individuals, RA: retirement-age (65+ years old) individuals. The numbers apply to the six-country average. 20 4 The impact of the Cost-of-Living Crisis In this section, we analyse the impact of the cost-of-living crisis on households in the Eurozone. In our baseline results, we study the magnitude of these effects on real wealth through the channels previously identified, separately for each income decile and age group. In a second step, in Section (5), we extend our analysis by considering various other subgroups, depending on characteristics such as liquid assets, housing and mortgage status. 4.1 Effects of Inflation Through Consumption, Income and Nominal Wealth We begin our analysis considering the impact of the inflationary surge on households across the income distribution through the relative consumption, nominal income and Fisher channels described above and summarised in Equ. (7). Table (3) summarises the main results for the Eurozone, as proxied by our six countries of interest (i.e. Germany, France, Italy, Spain, Greece and Portugal), which together represent about 85% of Eurozone GDP. Most importantly, the results point towards a striking difference in the extent of the exposure between working-age and retirement-age households. On average, pension-age households lost nearly three times as much as their working-age counterparts. This is mainly driven by the Fisher effect, which implies a devaluation of nominal wealth of about 8% in the case of pension-age households, who tend to hold larger stocks of positive net nominal assets. This is in sharp contrast to the revaluation of nominal wealth of about 4% for the group of working-age households, as a result of the fact that households in this group hold on average negative net nominal asset balances. Altogether, working-age households suffered an average loss from inflation amounting to ca. 6% of their disposable income, whereas the effect for those in pension-age amounted to 16%. These differences are consistent with the findings of the recent literature reviewed in the introduction section. Another important finding is the gap between high and low-income households. Our results suggest a regressive impact among working-age households, with the lowest deciles suffering between four to five times the impact borne by the highest deciles. In contrast, the impact appears rather flat among pension-age households, with households in this group experiencing losses of similar magnitudes across the income distribution. This result is related to differences in income and asset devaluations between age groups. On the nominal income side, pensions have grown relatively homogeneously across the income distribution following 21 the inflationary shock, whereas income growth was higher among high-income working-age households.17 Concerning the balance of nominal assets and liabilities, working-age households in high-income deciles are more likely to have positive mortgage balances and to benefit from the large debt devaluation from inflation. On the other hand, pensioners across all income deciles generally do not have mortgages and hold positive nominal asset balances, hence are loosing out by a similar magnitude (relative to their income) across the income distribution. Taken together, these factors are responsible for the significant difference in the regressivity of inflation effects between age groups. Finally, with respect to the inflation exposure resulting from decile-specific consumption patterns (the relative consumption channel), we find that households in the bottom income deciles, both in working and retirement-age groups, are more exposed than higher income households to price increases on goods such as fuel and electricity, that have featured above-average price rises, as documented in other studies (see, e.g., Amores et al.,2023b). In monetary terms, this is equivalent to a loss of approximately 1.9% and 1.7%, respectively, of disposable income for households in the first income decile, as compared to the country average. The consumption channel is noticeably smaller in magnitude than the Fisher and income channels. Figure (1) shows the effects of inflation by income and age group for the Eurozone as a whole and for each country separately. As it can be seen from the figure, the differences in the effects of inflation across age and income groups discussed above are present in all countries: pensionage households tend to experience significantly larger losses than their working-age counterparts. Moreover, among working-age households, the impact of inflation appears regressive, with low-income households suffering the greatest losses. Across countries, differences in the distribution of net nominal asset positions are the key driver of the differences in the magnitude and the regressive nature of the impact of inflation. For instance, comparing the case of German and Greek pensioners, it can be seen that the devaluation of large asset balances led to a significant loss of beyond 20% of disposable income for the German middle-income pensionage households, whereas Greek pension-age households suffered losses which are only about half of that size, due to their smaller nominal asset positions. At the other extreme, negative net nominal asset positions among higher-income working-age households in France and Spain imply that these households even benefited from the inflationary shock. 17This fact can be appreciated in Figure A.2 in the Appendix of this paper, where we plot the value of λ(from Equ. (2)) by age and income groups in each country. 22 their large asset balances which are positively exposed to higher interest rates. Altogether, at the Eurozone level, we observe a regressive impact of the interest rate response.21 At the country level, Figure (3) suggests that in most countries the URE tends to gradually increase along the income distribution and it is positive for pension-age households. Hence the impact of interest rate increases is regressive in all countries, with low-income workingage households typically suffering from interest rate increases, whereas high-income pensionage households are the main winners. There is, however, a substantial degree of variation across countries in the magnitude of the impact. The largest negative effects are suffered by households in the first income decile in Spain (working-age only) and Greece (both age groups) experiencing a loss of 5% to 7% of disposable income, while the impact is close to zero for lowincome working-age households in Germany. On the other hand, the highest income pensionage households in Spain are the ones benefiting the most from increases in rates thanks to a large stock of wealth (relative to their income) exposed to the higher interest rate. Looking at the decomposition of the URE by country (see Figure (A.6) in the Appendix) and income decile, consumption in excess of income (i.e. negative net savings) is often responsible for the bigger share of the URE in the first decile and particularly so in Greece, Spain and Italy. Moreover, cross-country differences in the URE – and by this account with respect to the impact of rising interest rates – can be traced back to the prevalence of adjustable-rate mortgage types in some countries as opposed to fixed-rate ones (see in particular Portugal and Spain). 21We should however be cautious in the interpretation of this result, as our analysis abstracts from the effect of interest rate hikes on inflation. It is therefore likely that households suffering the most from the direct effects of interest rates have benefited to a large extent from the fact that the monetary policy response prevented inflation from increasing further. Studying the impact of those second-round effects goes beyond the scope of this paper; we leave this for further research. 29 TABLE 5: Interest Rate Impact by Income Decile for Eurozone Households in 2022 Unhedged Interest Financial gain/loss Income decile Rate Exposure (URE) from interest rate hike (in % of disp. income) (in % of disp. income) Working-age 1−40.76 −1.83 2−14.42 −0.65 3−24.05 −1.08 4−5.35 −0.24 5−4.78 −0.22 6−1.61 −0.07 7 1.87 0.08 8 11.70 0.53 9 22.85 1.03 10 40.54 1.82 Pension-age 1 18.57 0.84 2 30.07 1.35 3 35.47 1.60 4 53.04 2.39 5 52.66 2.37 6 70.87 3.19 7 84.19 3.79 8 84.03 3.78 9 101.47 4.57 10 124.17 5.59 Notes: The table reports for each income decile the exposure to interest rate changes, as measured by the unhedged interest rate exposure (URE) in % of disposable income, and the actual financial impact resulting from the interest rate hike in % of disposable income. The actual monetary impact from the interest rate hike between mid-2021 and end-2023 is obtained by multiplying the decile-specific URE by 4.5% (see Equ. (9)). A negative number signals a negative exposure to rising interest rates, while positive ones indicate gains. The figures are the weighted average of six countries, which are France, Germany, Greece, Italy, Portugal and Spain. 30 FIGURE 3: Interest Rate Impact on Households Across the Eurozone (A) Eurozone -2.5 0 2.5 5 7.5 12345678910 12345678910 Working Age Pension Age in % of disposable income (B) France -10 -5 0 5 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (C) Germany -10 -5 0 5 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (D) Greece -10 -5 0 5 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (E) Italy -10 -5 0 5 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (F) Portugal -10 -5 0 5 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (G) Spain -10 -5 0 5 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income Notes: The figure shows for each income decile the monetary loss from an increase in the interest rate by 4.5%. Panel (3A) shows the weighted average effects across the six selected countries. 31 4.3 Relative gains and losses from the cost-of-living crisis This section brings together the analysis of the direct effects of inflation and the impact from fiscal and monetary policy responses, to assess the overall impact of the cost-of-living crisis on Euro Area households. The combined results are displayed in Figure (4) both for the Eurozone as a whole and for the individual countries. Starting with the Eurozone as a whole (Panel (4A)), our results indicate that the regressive direct effects of inflation prevail among the working-age population. While this was partially dampened by supporting fiscal policy measures, the rise in interest rates has reinforced the regressive effects of the crisis for this population group. Overall, the loss for working-age households ranges from 10% of disposable income for low-income households to virtually zero for high-income households. In contrast, among pension-age households, the impact was visibly less regressive in nature, and although the increase in rates had a mostly beneficial effect, the large devaluation of nominal balances induced by inflation implied that, across the income distribution, pension-age households were the most affected, with a loss of beyond 10%. At the country level, our results indicate that fiscal policy measures in Portugal have entirely offset the effect of inflation on working-age households. By the end of 2023, Portuguese households in this age group were in most cases fully compensated. In contrast, regressive patterns are still visible in most countries among working-age households with low-income households experiencing losses of about 8% to 10% of disposable income in Germany, Italy and Spain, whereas high-income households are equally well or even better off, as in the case of France. As for pension-age households, they tend to display similar losses across income groups and countries of about and beyond 10%. 32 FIGURE 4: The Impact of the Cost-of-Living Crisis on Households in the Eurozone (A) Eurozone -30 -20 -10 0 10 20 30 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Fisher effect Nom. income channel Rel. consumption channel Fiscal measures Monetary measures Total effect Working Age Pension Age in % of disposable income (B) France -30 -20 -10 0 10 20 30 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (C) Germany -30 -20 -10 0 10 20 30 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (D) Greece -30 -20 -10 0 10 20 30 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (E) Italy -30 -20 -10 0 10 20 30 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (F) Portugal -30 -20 -10 0 10 20 30 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income (G) Spain -30 -20 -10 0 10 20 30 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Working Age Pension Age in % of disposable income Notes: The figure shows for each income decile the monetary loss from inflation relative to disposable income through a revaluation of nominal assets (Fisher effect), nominal income and consumption, together with the effect resulting from fiscal and monetary responses. Panel (4A) shows the weighted average effects across the six selected countries. 33 5 The Role of Wealth and its Composition In the previous section, we analysed the effects of the cost-of-living crisis on various age and income subgroups of Euro Area households. We now extend the analysis by looking at how additional characteristics influence the exposure of households to inflation and the policy response, focusing on characteristics related to wealth and its composition. In particular, we look at the effects of home ownership, the mortgage status and the ‘hand-to-mouth’ status of households, as proxied by their levels of liquid wealth holdings. Housing In Section (4), we reported large differences in the effects of inflation through the Fisher channel across age groups. We found that, on average, working-age households benefit from inflation through this channel because of negative net nominal positions, while pensionage households lose from the devaluation of their nominal assets. In Figure (5), we depict the effects of the crisis as a function of the home-ownership status for each age-income decile group. We can see from the figure that, when accounting for home ownership, the gains from the Fisher effect within the working-age population (left panels) are driven by the effect of the shock on homeowners, who benefit the most from inflation. Those effects are non-monotonic over income deciles: the home-owners benefiting the most from the revaluation of nominal liabilities are those in middle-income deciles, the effect being as high as 10% of disposable income in the 7th income decile. Home-owners in low-income deciles appear to gain very little, and even lose in the case of the first decile. Non-homeowners have on average positive net nominal asset positions in all income deciles. As a result, working-age households in this group lose from inflation through the Fisher effect, as is the case for the retirement-age group. Looking at the pension-age population (right panels), the effects of inflation are very homogeneous across home-ownership groups. This makes the home-ownership status a poor predictor of the effects of the crisis on pension-age individuals. 34 FIGURE 5: Effects of Cost-of-Living Crisis Across Population Subgroups and Housing Status -20 -10 0 10 20 12345678910 12345678910 12345678910 12345678910 Non-Home Owners Home Owners Non-Home Owners Home Owners Fisher effect Nom. income channel Rel. consumption channel Fiscal measures Monetary measures Total effect Working Age Pension Age in % of disposable income Notes: The figure shows the monetary loss from inflation as a share of disposable income through a revaluation of nominal assets (Fisher effect), nominal income and consumption, together with the effect resulting from fiscal and monetary responses. The figure shows the weighted average effects across the six selected countries. Mortgages The stark difference in the role played by home ownership across age groups can be explained by accounting for the mortgage status of individuals more explicitly. As it can be seen from looking at Table (2), only about 8% of the retirement-age population has positive mortgage balances. For the working-age population, this share is equal to 60%. Moreover, conditional on having a mortgage, the average value of the remaining mortgage balance is equal to 198% of disposable income for working-age households, and to 156% for their retirement-age counterpart. Figure (6) displays the effects of the crisis on households conditional on their mortgage status.22 From the figure, we clearly see that mortgage holdings are a strong predictor of the exposure to inflation through the Fisher channel. For mortgage holders, the gains from inflation comove negatively with income, with low-income households experiencing the largest gains as a fraction of their disposable income. The effects are strong in magnitude, with mortgagees in the first decile seeing their net wealth revalued upwards by approximately 35%. 22The very low reported share of retirement-age households with a mortgage implies very noisy numbers when computing the effects on this group of individuals. Therefore, the figure only displays results for retirement-age households without mortgages. 35 Conditional on having positive mortgage balances, we show in the Appendix the effects of having a fixed-rate vs. adjustable-rate mortgage contract. We find that gains from the Fisher effect are similar across households with fixed and adjustable-rate contracts. However, the gains for households with adjustable-rate mortgages are partially offset by the losses they face from higher interest rates, given their large negative URE, reflecting the fact that interest payments on mortgages with adjustable rate increase following a monetary tightening. Given the strong cross-country differences in the type of the average mortgage contract (as can be seen from the country-specific population distribution tables in Section (C.1) of the Appendix),23 losses associated with having an adjustable-rate mortgage are highly concentrated in countries such as Spain and Portugal. We present country-specific results on the effects across housing and mortgage groups in the Appendix (Figures (A.3) and (A.4)). Overall, we find that the results at the country level feature striking similarities: the group of working-age home-owners, and in particular those with outstanding mortgages, have seen their wealth being devalued least or even gained, as is the case of mortgage holders. On the other hand, households without mortgages, both in working and pension age, have experienced wealth losses of very similar magnitudes. In terms of differences across the six countries, we notice somewhat smaller negative wealth effects on mortgage holders in France and Germany from the interest rate response, given the dominance of fixedrate mortgage regimes (cf. Section (C.1) for the country-specific shares of mortgage types by income decile). 23Those differences have also been documented in the literature, see e.g. Pica (2021). 36 FIGURE 6: Effects of the Cost-of-Living Crisis Across Population Subgroups and Mortgage Status -20 0 20 40 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 No Mortgage Mortgage No Mortgage Fisher effect Nom. income channel Rel. consumption channel Fiscal measures Monetary measures Total effect Working Age Pension Age in % of disposable income Notes: The figure shows the monetary loss from inflation as a share of disposable income through a revaluation of nominal assets (Fisher effect), nominal income and consumption, together with the effect resulting from fiscal and monetary responses. The figure shows the weighted average effects across the six selected countries. Hand-to-mouth status In Figure (7) we split the population into households considered as being ‘Hand-to-Mouth’ (i.e. holding no or little amounts of liquid assets) or not. The lack of liquid wealth exposes households to fluctuations in their earnings, implying that they typically have a high marginal propensity to consume out of transitory earnings shocks. Knowing whether hand-to-mouth households are more affected by the crisis is therefore of interest for the design of the policy response to the inflationary shock. Within the group of HtM households, we further distinguish between those with positive illiquid wealth, the ‘wealthy HtM’ households (typically owning housing wealth financed with mortgage debt) and the ‘poor HtM’ with no illiquid wealth.24 The results closely mirror those obtained looking at the housing and mortgage status: non-HtM households have been negatively affected by inflation through a devaluation of their nominal assets. Poor HtM consumers are protected from the Fisher effect as they hold, by definition, no net nominal assets. The total effect, however, is negative and of similar magnitude in both cases. Wealthy HtM consumers, 24This classification follows the influential work of Kaplan and Violante (2014). 37 FIGURE 7: Effects of the Cost-of-Living Crisis Across Population Subgroups and Hand-toMouth Status -20 0 20 40 12345678910 12345678910 12345678910 12345678910 12345678910 12345678910 Non-HtM Poor HtM Wealthy HtM Non-HtM Poor HtM Wealthy HtM Fisher effect Nom. income channel Rel. consumption channel Fiscal measures Monetary measures Total effect Working Age Pension Age in % of disposable income Notes: The figure shows the monetary loss from inflation as a share of disposable income through a revaluation of nominal assets (Fisher effect), nominal income and consumption, together with the effect resulting from fiscal and monetary responses. The figure shows the weighted average effects across the six selected countries. on the other hand, benefit from a substantial devaluation of their nominal liabilities, leaving those households with an overall gain of around 8% of disposable income on average. Net worth Figure 8depicts the impact of the cost-of-living crisis along deciles of the net wealth distribution (rather than disposable income as in the baseline results). Most of the variations in the effects of inflation across wealth groups can be explained through differences in exposures through the revaluation of net nominal positions (the Fisher effect). Comparing the effects for the two age groups, results are similar to those presented in Figure (4): pensionage households face on average substantial losses through the Fisher effect, while working-age households tend to benefit from it. However, it can be seen from the figure that some pensionage individuals in low-wealth deciles gain from the Fisher effect, due to their negative nominal asset positions. For pension-age households, losses from inflation increase with the level of wealth, as wealthier individuals in this age group suffer more from the devaluation of their nominal asset holdings, which are increasing with wealth. For working-age households, the effects are non-monotonic across net wealth deciles, with middle-wealth households suffering 38 Fiscal impact We consider the effects of changes in discretionary measures (“income-side measures”) ˜ Tj,t, together with changes in consumption taxes (“price-side measures”) on individual goods, τc k,t. Using 1 +πk,t=1+τc k,t 1+τc k,t−1 Pk,t Pk,t−1≈τc k,t−τc k,t−1+˜ πk,t, 1 +πt=1+τc t 1+τc t−1 Pt Pt−1≈ τc t−τc t−1+˜ πt, we get the following expression from differentiating (A.1) wrt τc k,t,τc tand ˜ Tj,t: da(τ) j,t=d˜ Tj,t+cj.t∑ k ωj,k,tdτc k−dτc+∑ s≥0 Q(t+s) tB(t+s) j,t−1+ (1−λj,t)y(t) j,t−1dτc. Re-arranging and using the NNP definition, we get equation (8) in the main text. Interest rate impact To obtain da(R) j,t, first notice that, given our assumptions, all bond prices move equally by the amount dQ(t+s) Q(t+s)=dq(t+s) q(t+s)=−dR Rfor all s≥1. To compute the effect of a change in interest rates, we consider a change in the value of goods today in tomorrow’s terms (Q(t) t, which so far was normalised to one), rather than the change in tomorrow goods in today’s term.26 To do so, we consider the effect of a dR increase in Q(t) t (so far normalised to ones) rather than a dR decrease in Q(t+s) tfor s≥1, which we normalise to one. In this case, the household budget constraint (A.1) can be written as: aj,t=Q(t) t"yj,t+b(t) j,t−1+1 1+πt B(t) j,t−1+˜ Tj,t−cj,t∑ k ωj,k,t 1+πk,t 1+πt# +1 1+πt∑ s≥1 Q(t+s) tB(t+s) j,t−1+∑ s≥1 (1+π)sq(t+s) tb(t+s) j,t−1 =Q(t) tUREj,t+1 1+πt∑ s≥1 Q(t+s) tB(t+s) j,t−1+∑ s≥1 (1+π)sq(t+s) tb(t+s) j,t−1 where aj,tis now expressed in terms of today’s goods Q(t) t. From this equation, we get: da(R) j,t=UREj,tdQ(t)=UREj,tdR which is the equation stated in the main text. 26Auclert (2019) uses a similar argument to compute the effects of monetary policy on household consumption. 45 B Data and Empirical Construction of Variables B.1 Computing the Net Nominal Asset Position (NNP) We follow the approach in Doepke and Schneider (2006) and Pallotti et al. (2023) and define the “Net Nominal Asset Position” (NNP) as the difference between the sum of nominal assets, comprising deposits, bonds and money owned to the household, and the sum of liabilities. Liabilities include both mortgage debt and non-mortgage debt (credit lines, credit cards and other non-collateralized loans). Table (A.1) provides details on the specific variables that were used to construct the NAP based on HFCS data. TABLE A.1: Construction of the Net Nominal Asset Position (NNP) from HFCS Data HFCS Variable Description Nominal assets HD1110 Value of sight account HD1210 Value of saving accounts DA2103 Bonds HD1701 Money owed to households Nominal liabilities DL1110 Outstanding balance of households’ main residence mortgages DL1120 Outstanding balance of mortgages on other properties DL1210 Outstanding balance of credit line/overdraft DL1220 Outstanding balance of credit card debt DL1231 Outstanding balance of private loans DL1232 Outstanding balance of other non-private non-collateralised loans Notes: The variable names refer to the third wave of the Household Finance and Consumption Survey (HFCS). B.2 Computing the Unhedged Interest Rate Exposure (URE) The following table provides details on the specific variables that were used to construct the URE. This approach follows closely the elaborations in Tzamourani (2021). 46 TABLE A.2: Construction of Components of the Unhedged Interest Rate Exposure (URE) HFCS Variable Description Adjustment Net Income DI2000 Total household gross income Net income obtained from net-to-gross income ratios from EUROMOD (2023) Consumption Consumption-to-net-income ratios obtained from HBS by country and income decile, applied to net income above. HB2300 (Monthly) amount paid as rent ×12 to obtain annual value Liabilities DL1110a Outstanding balance of adjustable interest rate HMR mortgages DL1120a Outstanding balance of adjustable interest rate mortgage on other properties DL1200 Outstanding balance of other, non-mortgage debt HB170x, x={1, 2, 3}Fixed rate mortgage 1, 2 or 3 on household’s main residence with maturity of 1 year or less (HB171x ≤1) HB370xy, x,y={1, 2, 3}Other fixed rate mortgage 1, 2 or 3 on household’s other properties 1, 2 or 3 with maturity of 1 year or less (HB371xy ≤1) Assets HD1110 Value of sight accounts HD1210 Value of saving accounts ×0.8 HD1320b Value of mutual funds invested in bonds ×0.9 HD1320c Value of mutual funds invested in money market ×0.9 HD1420 Value of bonds Multiplied with respective share by country, see Tzamourani (2021, Table A1) HD1620 value of additional assets in managed accounts ×0.9 Notes: All variable names refer to the third wave of the Household Finance and Consumption Survey (HFCS). 47 B.3 Identifying Hand-to-Mouth Households For the classification of households into HtM status we follow Almgren et al. (2022) and Kaplan et al. (2014). A household is classified as HtM if its net balance of liquid wealth is smaller than a certain share of monthly income. Following the authors’ notation, let midenote net liquid assets, yidenote income, and mibe a credit limit for household i, which is set to be the household’s monthly income. Then, a household is categorized as HtM if 0≤mi≤yi 2, or if 0≤mi, and mi≤yi 2−mi. Within the group of HtM households, we distinguish between ‘wealthy’ and ‘poor’. Wealthy HtM have a positive net illiquid wealth balance, while poor HtM have zero or negative net illiquid wealth balances. 48 C Additional Tables C.1 Population Distribution Across Income Deciles in Individual Countries France TABLE A.3: Population Distribution Across Income Deciles in France Decile 1 2 3 4 5 6 7 8 9 10 A All WA 7.82% 5.91% 6.12% 6.62% 6.28% 6.68% 7.14% 7.54% 7.87% 8.03% RA 2.19% 4.09% 3.94% 3.31% 3.75% 3.29% 2.86% 2.45% 2.14% 1.96% B Home-owner WA 2.04% 1.50% 1.91% 2.51% 3.09% 3.75% 5.06% 5.88% 6.94% 7.43% RA 1.26% 1.82% 2.28% 2.39% 3.01% 2.84% 2.58% 2.31% 1.89% 1.91% Non owner WA 5.78% 4.41% 4.21% 4.11% 3.19% 2.93% 2.09% 1.66% 0.92% 0.60% RA 0.94% 2.27% 1.66% 0.91% 0.74% 0.45% 0.28% 0.14% 0.25% 0.05% C Mortgage WA 0.50% 0.51% 0.93% 1.27% 1.79% 2.24% 3.20% 3.78% 4.43% 5.05% RA 0.04% 0.07% 0.08% 0.09% 0.15% 0.11% 0.25% 0.20% 0.22% 0.26% No mortgage WA 1.54% 0.99% 0.98% 1.24% 1.30% 1.51% 1.85% 2.10% 2.52% 2.38% RA 1.21% 1.75% 2.20% 2.30% 2.86% 2.73% 2.34% 2.11% 1.68% 1.65% Non owner WA 5.78% 4.41% 4.21% 4.11% 3.19% 2.93% 2.09% 1.66% 0.92% 0.60% RA 0.94% 2.27% 1.66% 0.91% 0.74% 0.45% 0.28% 0.14% 0.25% 0.05% D AR mortgage WA 0.04% 0.01% 0.07% 0.09% 0.10% 0.07% 0.19% 0.23% 0.22% 0.28% RA 0.00% 0.00% 0.00% 0.03% 0.02% 0.00% 0.00% 0.01% 0.01% 0.00% FR mortgage WA 0.46% 0.50% 0.85% 1.18% 1.69% 2.17% 3.02% 3.55% 4.20% 4.77% RA 0.04% 0.07% 0.08% 0.06% 0.14% 0.11% 0.25% 0.19% 0.20% 0.26% E Poor HTM WA 1.73% 1.68% 1.36% 1.17% 0.85% 0.80% 0.47% 0.33% 0.15% 0.04% RA 0.27% 0.57% 0.49% 0.25% 0.16% 0.14% 0.03% 0.03% 0.04% 0.00% Wealthy HTM WA 1.14% 0.82% 1.07% 1.41% 1.88% 2.11% 2.58% 2.84% 2.88% 2.83% RA 0.22% 0.37% 0.49% 0.36% 0.33% 0.25% 0.30% 0.15% 0.16% 0.14% Non HTM WA 4.94% 3.40% 3.69% 4.04% 3.55% 3.77% 4.10% 4.38% 4.84% 5.16% RA 1.70% 3.15% 2.96% 2.70% 3.26% 2.91% 2.52% 2.27% 1.94% 1.82% F Effective πWA 15.07% 15.01% 14.97% 15.06% 14.93% 14.89% 14.81% 14.65% 14.49% 14.41% RA 16.64% 16.48% 16.45% 16.14% 16.04% 15.88% 15.91% 15.77% 15.16% 15.24% Notes: WA: Working-age (<65 years old) individuals, RA: Retirement-age (65+ years old) individuals. 49 Germany TABLE A.4: Population Distribution Across Income Deciles in Germany Decile 1 2 3 4 5 6 7 8 9 10 A All WA 7.22% 5.21% 5.58% 6.33% 6.63% 7.14% 7.87% 8.49% 8.67% 8.66% RA 2.79% 4.80% 4.43% 3.70% 3.37% 2.81% 2.13% 1.56% 1.28% 1.32% B Home-owner WA 0.56% 0.74% 1.64% 2.00% 2.57% 3.05% 4.18% 5.65% 6.34% 7.29% RA 0.78% 1.35% 2.30% 2.29% 1.86% 2.06% 1.56% 1.24% 1.09% 1.22% Non owner WA 6.66% 4.46% 3.94% 4.32% 4.06% 4.09% 3.69% 2.84% 2.33% 1.37% RA 2.01% 3.44% 2.14% 1.41% 1.51% 0.74% 0.57% 0.32% 0.19% 0.11% C Mortgage WA 0.16% 0.15% 0.57% 0.56% 1.11% 1.80% 2.42% 3.05% 4.20% 4.57% RA 0.09% 0.10% 0.15% 0.28% 0.27% 0.27% 0.15% 0.27% 0.32% 0.37% No mortgage WA 0.40% 0.59% 1.07% 1.44% 1.45% 1.25% 1.76% 2.60% 2.14% 2.72% RA 0.69% 1.26% 2.14% 2.02% 1.59% 1.79% 1.40% 0.97% 0.77% 0.84% Non owner WA 6.66% 4.46% 3.94% 4.32% 4.06% 4.09% 3.69% 2.84% 2.33% 1.37% RA 2.01% 3.44% 2.14% 1.41% 1.51% 0.74% 0.57% 0.32% 0.19% 0.11% D AR mortgage WA 0.06% 0.02% 0.25% 0.04% 0.29% 0.34% 0.25% 0.56% 0.50% 0.57% RA 0.02% 0.00% 0.02% 0.03% 0.01% 0.00% 0.03% 0.02% 0.01% 0.04% FR mortgage WA 0.10% 0.13% 0.32% 0.52% 0.82% 1.46% 2.17% 2.49% 3.70% 4.00% RA 0.07% 0.09% 0.14% 0.25% 0.26% 0.27% 0.13% 0.26% 0.30% 0.33% E Poor HTM WA 1.95% 1.56% 1.03% 0.99% 0.42% 0.49% 0.51% 0.30% 0.07% 0.07% RA 0.54% 1.31% 0.55% 0.21% 0.22% 0.08% 0.03% 0.05% 0.00% 0.00% Wealthy HTM WA 0.36% 0.74% 1.27% 1.29% 2.13% 1.90% 1.96% 2.49% 2.37% 2.37% RA 0.17% 0.11% 0.38% 0.28% 0.30% 0.25% 0.09% 0.14% 0.08% 0.15% Non HTM WA 4.91% 2.91% 3.29% 4.04% 4.07% 4.75% 5.39% 5.70% 6.24% 6.23% RA 2.08% 3.38% 3.50% 3.21% 2.85% 2.48% 2.00% 1.38% 1.20% 1.17% F Effective πWA 18.72% 18.05% 17.69% 17.53% 17.39% 17.30% 17.29% 17.21% 17.01% 16.54% RA 18.98% 18.48% 18.08% 17.99% 17.81% 17.58% 17.39% 16.82% 16.39% 15.75% Notes: WA: Working-age (<65 years old) individuals, RA: Retirement-age (65+ years old) individuals. 50 Greece TABLE A.5: Population Distribution Across Income Deciles in Greece Decile 1 2 3 4 5 6 7 8 9 10 A All WA 6.64% 4.75% 5.88% 5.54% 6.55% 5.94% 6.61% 7.32% 7.84% 8.30% RA 3.37% 5.25% 4.13% 4.47% 4.06% 3.52% 3.33% 2.66% 2.15% 1.69% B Home-owner WA 3.59% 2.84% 3.21% 3.33% 4.11% 4.77% 4.89% 6.22% 6.27% 7.84% RA 2.61% 4.33% 3.57% 4.15% 3.69% 3.36% 3.31% 2.62% 2.11% 1.58% Non owner WA 2.84% 1.92% 2.68% 2.23% 2.46% 1.19% 1.74% 1.12% 1.59% 0.48% RA 0.74% 0.93% 0.57% 0.33% 0.38% 0.16% 0.03% 0.05% 0.04% 0.12% C Mortgage WA 0.85% 0.34% 0.71% 0.71% 0.68% 0.67% 0.86% 1.58% 1.05% 1.84% RA 0.09% 0.08% 0.07% 0.15% 0.23% 0.16% 0.04% 0.18% 0.11% 0.31% No mortgage WA 2.74% 2.50% 2.50% 2.62% 3.43% 4.10% 4.02% 4.64% 5.22% 6.00% RA 2.52% 4.25% 3.50% 4.00% 3.46% 3.20% 3.27% 2.44% 2.00% 1.26% Non owner WA 2.84% 1.92% 2.68% 2.23% 2.46% 1.19% 1.74% 1.12% 1.59% 0.48% RA 0.74% 0.93% 0.57% 0.33% 0.38% 0.16% 0.03% 0.05% 0.04% 0.12% D AR mortgage WA 0.39% 0.16% 0.44% 0.39% 0.34% 0.36% 0.49% 0.74% 0.58% 1.01% RA 0.07% 0.03% 0.03% 0.05% 0.04% 0.09% 0.04% 0.07% 0.00% 0.06% FR mortgage WA 0.47% 0.17% 0.27% 0.32% 0.34% 0.32% 0.37% 0.84% 0.47% 0.84% RA 0.02% 0.05% 0.04% 0.11% 0.19% 0.07% 0.00% 0.10% 0.11% 0.25% E Poor HTM WA 1.59% 1.20% 1.87% 1.45% 1.26% 0.55% 1.16% 0.57% 0.68% 0.17% RA 0.38% 0.52% 0.20% 0.25% 0.17% 0.09% 0.00% 0.01% 0.04% 0.00% Wealthy HTM WA 2.42% 1.68% 1.71% 1.35% 1.86% 1.77% 2.31% 2.89% 2.63% 2.91% RA 1.84% 2.32% 1.48% 1.84% 1.53% 1.33% 0.83% 0.89% 0.81% 0.39% Non HTM WA 1.13% 2.42% 2.45% 2.40% 2.37% 2.11% 2.51% 1.76% 1.31% 1.31% RA 3.72% 1.88% 3.57% 3.03% 3.51% 3.56% 3.79% 4.33% 4.41% 5.44% F Effective πWA 16.49% 16.12% 16.14% 16.05% 15.87% 15.70% 15.45% 15.30% 15.12% 14.51% RA 17.52% 17.42% 17.09% 17.06% 16.77% 16.39% 16.29% 16.22% 15.93% 15.56% Notes: WA: Working-age (<65 years old) individuals, RA: Retirement-age (65+ years old) individuals. 51 Italy TABLE A.6: Population Distribution Across Income Deciles in Italy Decile 1 2 3 4 5 6 7 8 9 10 A All WA 6.33% 4.11% 5.36% 6.06% 6.50% 6.40% 6.80% 7.23% 7.62% 7.76% RA 3.68% 5.93% 4.60% 3.94% 3.50% 3.61% 3.20% 2.80% 2.35% 2.23% B Home-owner WA 2.65% 1.44% 2.72% 3.10% 3.60% 4.29% 4.99% 5.79% 6.58% 7.26% RA 2.30% 4.08% 3.25% 3.07% 2.98% 2.87% 2.82% 2.61% 2.18% 2.06% Non owner WA 3.67% 2.67% 2.63% 2.96% 2.90% 2.11% 1.81% 1.44% 1.04% 0.50% RA 1.38% 1.85% 1.35% 0.87% 0.52% 0.74% 0.38% 0.19% 0.17% 0.17% C Mortgage WA 0.28% 0.10% 0.68% 0.42% 0.40% 0.51% 1.02% 1.10% 1.56% 2.30% RA 0.01% 0.04% 0.03% 0.03% 0.06% 0.06% 0.07% 0.11% 0.05% 0.09% No mortgage WA 2.37% 1.34% 2.04% 2.68% 3.20% 3.78% 3.97% 4.69% 5.03% 4.96% RA 2.29% 4.05% 3.21% 3.04% 2.92% 2.81% 2.76% 2.50% 2.13% 1.97% Non owner WA 3.67% 2.67% 2.63% 2.96% 2.90% 2.11% 1.81% 1.44% 1.04% 0.50% RA 1.38% 1.85% 1.35% 0.87% 0.52% 0.74% 0.38% 0.19% 0.17% 0.17% D AR mortgage WA 0.06% 0.09% 0.36% 0.17% 0.26% 0.29% 0.51% 0.58% 0.79% 1.29% RA 0.01% 0.00% 0.01% 0.00% 0.02% 0.00% 0.03% 0.04% 0.01% 0.01% FR mortgage WA 0.22% 0.01% 0.32% 0.25% 0.14% 0.22% 0.51% 0.52% 0.77% 1.01% RA 0.00% 0.04% 0.03% 0.03% 0.04% 0.05% 0.04% 0.06% 0.05% 0.08% E Poor HTM WA 2.74% 1.42% 0.91% 0.76% 0.44% 0.24% 0.42% 0.12% 0.15% 0.06% RA 0.45% 0.58% 0.43% 0.19% 0.12% 0.19% 0.13% 0.01% 0.12% 0.00% Wealthy HTM WA 1.62% 0.62% 1.23% 1.13% 0.84% 1.02% 1.41% 1.04% 1.38% 1.52% RA 1.02% 0.98% 0.52% 0.30% 0.43% 0.22% 0.19% 0.25% 0.25% 0.13% Non HTM WA 1.97% 2.06% 3.21% 4.17% 5.22% 5.14% 4.97% 6.06% 6.09% 6.16% RA 2.21% 4.38% 3.66% 3.45% 2.95% 3.20% 2.89% 2.54% 1.98% 2.10% F Effective πWA 20.84% 18.56% 18.03% 17.73% 17.10% 16.53% 16.38% 15.75% 15.33% 14.74% RA 21.15% 19.90% 19.15% 18.62% 18.28% 17.71% 16.76% 16.68% 16.14% 15.63% Notes: WA: Working-age (<65 years old) individuals, RA: Retirement-age (65+ years old) individuals. 52 Portugal TABLE A.7: Population Distribution Across Income Deciles in Portugal Decile 1 2 3 4 5 6 7 8 9 10 A All WA 4.80% 4.21% 5.72% 5.49% 7.51% 7.27% 8.02% 8.40% 8.21% 8.17% RA 5.22% 5.78% 4.33% 4.45% 2.54% 2.74% 1.92% 1.61% 1.80% 1.82% B Home-owner WA 2.24% 2.40% 3.69% 3.85% 5.32% 5.67% 6.57% 7.56% 7.13% 7.76% RA 3.78% 4.12% 3.37% 3.66% 2.30% 2.44% 1.79% 1.53% 1.61% 1.71% Non owner WA 2.57% 1.81% 2.03% 1.65% 2.19% 1.60% 1.46% 0.85% 1.07% 0.41% RA 1.44% 1.66% 0.96% 0.79% 0.24% 0.30% 0.13% 0.08% 0.19% 0.11% C Mortgage WA 0.80% 0.89% 2.00% 1.96% 3.15% 3.63% 3.99% 5.31% 4.88% 5.50% RA 0.15% 0.16% 0.16% 0.17% 0.08% 0.17% 0.17% 0.30% 0.30% 0.33% No mortgage WA 1.44% 1.50% 1.69% 1.89% 2.17% 2.03% 2.58% 2.25% 2.25% 2.26% RA 3.63% 3.96% 3.21% 3.48% 2.23% 2.28% 1.62% 1.23% 1.31% 1.38% Non owner WA 2.57% 1.81% 2.03% 1.65% 2.19% 1.60% 1.46% 0.85% 1.07% 0.41% RA 1.44% 1.66% 0.96% 0.79% 0.24% 0.30% 0.13% 0.08% 0.19% 0.11% D AR mortgage WA 0.76% 0.84% 1.81% 1.70% 2.54% 3.20% 3.16% 4.76% 4.37% 4.47% RA 0.13% 0.11% 0.10% 0.15% 0.05% 0.06% 0.12% 0.23% 0.15% 0.20% FR mortgage WA 0.04% 0.05% 0.19% 0.26% 0.62% 0.44% 0.83% 0.55% 0.51% 1.03% RA 0.03% 0.05% 0.06% 0.02% 0.03% 0.11% 0.04% 0.06% 0.15% 0.13% E Poor HTM WA 1.50% 0.87% 0.88% 0.63% 0.84% 0.49% 0.29% 0.22% 0.33% 0.02% RA 0.63% 0.69% 0.42% 0.17% 0.02% 0.07% 0.04% 0.02% 0.00% 0.00% Wealthy HTM WA 1.00% 1.33% 2.25% 1.68% 3.15% 3.35% 3.28% 3.91% 3.15% 2.72% RA 0.85% 0.92% 0.48% 0.40% 0.22% 0.60% 0.23% 0.14% 0.26% 0.15% Non HTM WA 2.30% 2.01% 2.59% 3.19% 3.51% 3.43% 4.45% 4.27% 4.73% 5.42% RA 3.75% 4.17% 3.44% 3.88% 2.30% 2.08% 1.65% 1.44% 1.54% 1.67% F Effective πWA 14.72% 14.73% 14.55% 14.55% 14.43% 14.21% 14.13% 13.99% 13.86% 13.66% RA 14.84% 15.19% 14.88% 15.04% 14.48% 14.59% 14.34% 14.27% 14.37% 13.82% Notes: WA: Working-age (<65 years old) individuals, RA: Retirement-age (65+ years old) individuals. 53 Spain TABLE A.8: Population Distribution Across Income Deciles in Spain Decile 1 2 3 4 5 6 7 8 9 10 A All WA 7.77% 3.89% 6.15% 6.15% 7.31% 6.82% 7.72% 7.75% 7.91% 8.27% RA 2.27% 6.07% 3.96% 3.90% 2.53% 3.18% 2.29% 2.36% 1.98% 1.72% B Home-owner WA 3.46% 2.75% 3.95% 4.46% 5.31% 5.77% 6.57% 6.97% 7.44% 7.59% RA 1.61% 4.92% 3.41% 3.35% 2.37% 2.95% 2.18% 2.28% 1.92% 1.71% Non owner WA 4.25% 1.15% 2.21% 1.69% 2.01% 1.06% 1.16% 0.79% 0.48% 0.69% RA 0.60% 1.16% 0.56% 0.56% 0.17% 0.23% 0.11% 0.08% 0.06% 0.01% C Mortgage WA 1.16% 1.16% 1.63% 2.42% 3.41% 3.30% 3.96% 4.43% 5.47% 4.77% RA 0.08% 0.21% 0.21% 0.25% 0.19% 0.23% 0.28% 0.22% 0.52% 0.43% No mortgage WA 2.30% 1.59% 2.32% 2.04% 1.91% 2.47% 2.61% 2.54% 1.97% 2.82% RA 1.52% 4.71% 3.20% 3.10% 2.17% 2.72% 1.90% 2.06% 1.40% 1.28% Non owner WA 4.25% 1.15% 2.21% 1.69% 2.01% 1.06% 1.16% 0.79% 0.48% 0.69% RA 0.60% 1.16% 0.56% 0.56% 0.17% 0.23% 0.11% 0.08% 0.06% 0.01% D AR mortgage WA 0.88% 0.80% 1.18% 1.94% 2.44% 2.59% 2.79% 3.26% 4.01% 3.03% RA 0.01% 0.08% 0.08% 0.09% 0.14% 0.06% 0.15% 0.10% 0.18% 0.06% FR mortgage WA 0.28% 0.35% 0.45% 0.48% 0.97% 0.70% 1.17% 1.17% 1.46% 1.74% RA 0.07% 0.12% 0.14% 0.16% 0.06% 0.17% 0.13% 0.12% 0.34% 0.37% E Poor HTM WA 2.22% 0.57% 0.94% 0.79% 0.70% 0.39% 0.33% 0.11% 0.02% 0.02% RA 0.30% 0.37% 0.18% 0.13% 0.07% 0.06% 0.01% 0.04% 0.02% 0.01% Wealthy HTM WA 1.76% 1.45% 1.65% 2.33% 3.11% 2.88% 3.61% 3.32% 3.49% 2.81% RA 0.35% 0.84% 0.58% 0.50% 0.27% 0.46% 0.27% 0.20% 0.26% 0.18% Non HTM WA 3.72% 1.88% 3.57% 3.03% 3.51% 3.56% 3.79% 4.33% 4.41% 5.44% RA 1.56% 4.87% 3.21% 3.28% 2.20% 2.67% 2.02% 2.12% 1.70% 1.53% F Effective πWA 13.59% 13.23% 12.87% 12.94% 12.75% 12.61% 12.73% 12.58% 12.52% 12.27% RA 14.09% 13.88% 13.51% 13.72% 13.48% 13.45% 13.28% 13.05% 12.95% 12.70% Notes: WA: Working-age (<65 years old) individuals, RA: Retirement-age (65+ years old) individuals. 54 FIGURE A.6: Decomposition of the Unhedged Interest Rate Exposure (URE) (A) Eurozone -50 0 50 100 150 12345678910 12345678910 Current period income minus consumption Net maturing asset position Maturing mortgage-related liabilities Total URE Working Age Pension Age in % of disposable income Total (B) France -150 -100 -50 0 50 100 150 12345678910 12345678910 Working Age Pension Age in % of disposable income FR (C) Germany -150 -100 -50 0 50 100 150 12345678910 12345678910 Working Age Pension Age in % of disposable income DE (D) Greece -150 -100 -50 0 50 100 150 12345678910 12345678910 Working Age Pension Age in % of disposable income EL (E) Italy -150 -100 -50 0 50 100 150 12345678910 12345678910 Working Age Pension Age in % of disposable income IT (F) Portugal -150 -100 -50 0 50 100 150 12345678910 12345678910 Working Age Pension Age in % of disposable income PT (G) Spain -150 -100 -50 0 50 100 150 12345678910 12345678910 Working Age Pension Age in % of disposable income ES Notes: The figure shows for each income decile and separately for the working and retirement-age the decomposition of the URE. Panel (A.6A) shows the weighted average across the six selected countries. The “Net maturing asset position” is defined as the net difference between the sum of all assets and non-mortgage-related liabilities. “Maturing mortgage-related liabilities” defines the subset of mortgage-related liabilities. See Table (A.2) for the exact HFCS variables and their definitions. 61 GETTING IN TOUCH WITH THE EU In person All over the European Union there are hundreds of Europe Direct centres. You can find the address of the centre nearest you online (european-union.europa.eu/contact-eu/meet-us_en). On the phone or in writing Europe Direct is a service that answers your questions about the European Union. You can contact this service: by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls), at the following standard number: +32 22999696, via the following form: european-union.europa.eu/contact-eu/write-us_en. FINDING INFORMATION ABOUT THE EU Online Information about the European Union in all the official languages of the EU is available on the Europa website (europeanunion.europa.eu). EU publications You can view or order EU publications at op.europa.eu/en/publications. Multiple copies of free publications can be obtained by contacting Europe Direct or your local documentation centre (european-union.europa.eu/contact-eu/meet-us_en). EU law and related documents For access to legal information from the EU, including all EU law since 1951 in all the official language versions, go to EUR-Lex (eur-lex.europa.eu). Open data from the EU The portal data.europa.eu provides access to open datasets from the EU institutions, bodies and agencies. These can be downloaded and reused for free, for both commercial and non-commercial purposes. The portal also provides access to a wealth of datasets from European countries.