Log in
E-mail
Password
Remember
Forgot password ?
Become a member for free
Sign up
Sign up
New member
Sign up for FREE
New customer
Discover our services
Settings
Settings
Dynamic quotes 
OFFON
News: Latest News
Latest NewsCompaniesMarketsEconomy & ForexCommoditiesInterest RatesBusiness LeadersFinance Pro.CalendarSectors 
All NewsEconomyCurrencies & ForexEconomic EventsCryptocurrenciesCybersecurityPress Releases

Do macroprudential measures increase inequality? Evidence from the euro area household survey

06/18/2021 | 05:15am EDT

Working Paper Series

Oana-Maria Georgescu, Diego Vila Martin Do macroprudential measures

increase inequality? Evidence from the euro area household survey

No 2567 / June 2021

Disclaimer: This paper should not be reported as representing the views of the European Central Bank (ECB). The views expressed are those of the authors and do not necessarily reflect those of the ECB.

Abstract

Borrower-based macroprudential (MP) policies - such as caps on loan-to-value (LTV) ratios and debt-service-to-income (DSTI) limits - contain the build-up of systemic risk by reducing the probability and conditional impact of a crisis. While LTV/DSTI limits can increase inequality at introduction, they can dampen the increase in inequality under adverse macroeconomic conditions. The relative size of these opposing eects is an empirical question. We conduct counterfactual simulations under dierent macroeconomic and macroprudential policy scenarios using granular income and wealth data from the Households Finance and Consumption Survey (HFCS) for Ireland, Italy, Netherlands and Portugal. Simulation results show that borrower-based measures have a moderate negative welfare impact in terms of wealth inequality and a negligible impact on income inequality.

JEL classication: G21, G28, G51.

Keywords: macroprudential policy, inequality, household debt.

ECB Working Paper Series No 2567 / June 2021

1

Non-technical Summary

This paper aims to asses the welfare costs and benets of borrower-based measures in terms of wealth and income inequality. Welfare costs arise if wealth and potentially income inequality increase after the imposition of LTV and DSTI limits compared to the counterfactual without these limits. The benets of borrower-based measures may occur as a result of a lower probability and conditional impact of a nancial crisis. Since unemployment in a recession is higher for low income borrowers, an adverse macroeconomic scenario with borrower-based measure in place may result in a lower increase in income inequality compared to the counterfactual adverse scenario without LTV and DSTI limits in place.

To quantify the size of these costs and benets, we conduct counterfactual simulations comparing wealth and income inequality under 4 policy scenarios (baseline and adverse scenario, with and without borrower-based measures in place). The paper restricts the scope of the study to four euro area countries: Ireland, Italy, Netherlands and Portugal.

To estimate the benets of macroprudential measures, we proceed in three steps. First, we use a Bayesian VAR (BVAR) to quantify the macroeconomic impact of a credit crunch conditional on an adverse macroeconomic scenario under dierent macroprudential policy regimes (i.e. with or without LTV and DSTI limits). In a second step, we match the unemployment and asset price response obtained in the BVAR to household data in order to derive income and wealth distributions under these dierent policy scenarios. Last, we compute the benet as the dierence in income (wealth) inequality between the adverse scenario without borrower-based macropru- dential measures compared to the adverse scenario with these measures in place. The cost is computed in a similar fashion, as the dierence in income and wealth inequality between the baseline scenarios with and without the borrower-based measures in place. The only dierence to the benets calculation is that the macroeconomic response is computed conditional on a loan demand shock following the implementation of borrower-based measures.

We use the Gini coecient of wealth and income as a metric for our cost-benet calculations. We derive the unconditional net benet as the dierence between the expected values of the Gini coecient with and without borrower-based measures, taking into account the change in the crisis probability across policy scenarios.

In terms of wealth inequality, the unconditional benet is negative, ranging from -0.14 percentage points in Italy to -0.58 percentage points in the Netherlands. The negative unconditional benet implies a net increase in wealth inequality relative to the counterfactual without borrower-based measures. In the case of income inequality, the unconditional benet of borrower-based measures is marginally positive, ranging from 0.04 percentage points for Italy to 0.16 percentage points for Ireland. These results suggest that the welfare cost of macroprudential regulation in terms of wealth inequality is contained, while the benet in terms of income inequality is negligible.

ECB Working Paper Series No 2567 / June 2021

2

  • Introduction

In many advanced economies, home ownership is part of the social contract between policy makers and citizens, acting as a symbol for social inclusion and economic growth. In this sense, the housing nance deregulation in the decade preceding the Great Financial Crisis in the US and UK was meant to act as a redistributive force, potentially reducing wealth and income inequality (Arundel and Ronald (2020), Rajan (2011), p. 35-40). After the bursting of the sub-prime housing bubble in 2008, access to mortgage loans has been tightened in most euro area countries. In this context, macroprudential policies (MP) have been widely used by policy makers due to their ability to address the build-up of asset price bubbles and excessive risk-taking by banks. Some of these measures can target banks, for example countercyclical capital requirements, while others target borrowers, such as the loan-to-value ratio (LTV) or the debt-service-to-income limit (DSTI).1

While the main policy objective of MP is to enhance nancial stability by increasing the resilience of banks, some studies pointed to the potential negative welfare eects of these policies in terms of wealth and income inequality (Frost and Stralen (2018); Carpantier et al. (2016)). In good times, LTV and DSTI limits may have direct redistributive eects by excluding low income households from the mortgage market. In bad times, macroprudential policy may dampen the increase in inequality by smoothing the credit cycle, thus lowering the probability and the conditional impact of a nancial crisis. The relative size of these opposing eects is an empirical question.

To quantify the size of these costs and benets, we conduct counterfactual simulations comparing wealth and income inequality under four policy scenarios (baseline and adverse scenario, with and without MP) for four euro area countries: Ireland, Italy, Netherlands and Portugal. The analysis combines granular household level data on wealth and income from the Euro Area Household Finance and Consumption Survey (HFCS) together with bank level data from the 2018 Euro Area Stress Test exercise and macroeconomic indicators from the ECB Statistical Data Warehouse.2 We use the Gini coecient as a metric for our cost-benet calculations.

To estimate the benets of macroprudential measures, we proceed in three steps. First, we use a Bayesian VAR (BVAR) to quantify the macroeconomic impact of a credit crunch conditional on an adverse macroeconomic scenario under dierent macroprudential policy regimes (i.e. with or without LTV and DSTI limits). In a second step, we match the unemployment and asset price response obtained in the BVAR to household data in order to derive income and wealth distributions under these dierent policy scenarios. Last, we compute the benet of the MP measures as the dierence in income (wealth) inequality between the adverse scenario without

1An LTV limit implies that a loan will be granted only if the borrower's down payment is suciently large relative to the value of the house. A DSTI limit excludes borrowers whose monthly mortgage payments exceed a certain portion of their monthly income.

  • In this paper we also used data from the DNB Household Survey.

ECB Working Paper Series No 2567 / June 2021

3

borrower-based MP measures and the adverse scenario with these measures in place. The cost is computed in a similar fashion, as the dierence in income (wealth) inequality between the baseline scenarios with and without these MP measures in place. The only dierence to the benets calculation is that the macroeconomic response is computed conditional on a loan demand shock at MP measures introduction.

Results show that the imposition of the DSTI and LTV limits results in an increase in the Gini coecient of wealth ranging from 0.14 percentage points in Italy to 0.58 percentage points in the Netherlands. In the baseline scenario with LTV and DSTI limits in place, two channels aect household wealth. First, the wealth of all households is aected by the impulse response of house, stock and bond prices to the loan demand shock implied by the borrower-based measures. Second, the counterfactual value of wealth of excluded households decreases by the current house value and increases by the initial mortgage amount. The change in wealth inequality is primarily driven by the change in wealth of excluded households after imposition of borrower-based limits. This welfare cost is expected given that (i) housing is the main source of wealth for households in the low percentiles of the wealth distribution and (ii) borrower-based limits are more binding for the latter households.

Somehow surprisingly, wealth inequality in the adverse scenario with borrower-based measures in place is higher than in the adverse scenario without these limits. While low-wealth households excluded from the mortgage market were shielded from a negative wealth shock following the fall in house prices in the adverse scenario, the decrease in wealth resulting from barriers to house ownership is larger. We derive the unconditional net benet as the dierence between the expected value of the Gini coecient with and without borrower-based measures, taking into account the change in the crisis probability across policy scenarios. The unconditional benet of borrower-based measures in terms of wealth inequality ranges from -0.14 percentage points in Italy to -0.58 percentage points in the Netherlands.

Turning to income inequality, a negligible decrease in the Gini coecient of income is observed after the imposition of LTV and DSTI limits, amounting to less than 0.04 percentage points in the baseline scenario. The small impact on income inequality is due to the fact that two opposing eects aect disposable income following the implementation of borrower-based limits. First, borrower-based limits have a contractionary impact at introduction, resulting in an increase in unemployment. Second, the debt burden decreases and therefore the disposable income of excluded households increases. These two eects cancel each other out, resulting in a muted impact on income inequality in the baseline scenario. In the adverse scenario income inequality remains broadly unchanged. The main reason for this result is the negligible change in the unemployment impulse response in the adverse scenario compared to the counterfactual with borrower-based measures in place. The unconditional benet of borrower-based measures in terms of income inequality is less than 0.04 percentage points of the Gini coecient of income.

ECB Working Paper Series No 2567 / June 2021

4

This is an excerpt of the original content. To continue reading it, access the original document here.

Disclaimer

ECB - European Central Bank published this content on 18 June 2021 and is solely responsible for the information contained therein. Distributed by Public, unedited and unaltered, on 18 June 2021 09:14:03 UTC.


ę Publicnow 2021
Latest news "Economy & Forex"
05:53pCorteva lifts sales forecast on strong demand for crop protection products, seeds
RE
05:51pLevi Strauss to acquire athletic wear maker Beyond Yoga
RE
05:45pCorteva lifts sales forecast on strong demand for crop protection products, seeds
RE
05:42pRICHARD BRANSON : Branson's Virgin Galactic to sell space flight tickets starting at $450,000
RE
05:39pEuro Lost 0.03% to $1.1835 -- Data Talk
DJ
05:39pSterling Gains 0.27% to $1.3928 -- Data Talk
DJ
05:39pDollar Gains 0.27% to 109.77 Yen -- Data Talk
DJ
05:34pUtilities Up Ahead Of Jobs Report -- Utilities Roundup
DJ
05:32pCommunications Services Up After Mixed Earnings -- Communications Services Roundup
DJ
05:30pTech Up After Chip Makers Address Shortages -- Tech Roundup
DJ
Latest news "Economy & Forex"