Contrary To Continued Claims QE Reduces Inequality

Quantitative easing has recently been shown to affect households differently depending on the composition of their income and wealth. Using euro area data, this column reviews the relevance of the direct and indirect effects of monetary policy on households’ incomes, which varies depending on employment status. The indirect income channel is found to be quantitatively more powerful, and especially beneficial for households holding few or no liquid assets. This implies that expansionary monetary policy in the euro area has led to a reduction in inequality.

The impact of monetary policy on inequality has recently come to public attention. For example, commentators have pointed out that a prolonged reduction in interest rates generates an income loss for savers holding interest-bearing assets, while it benefits younger households that are net borrowers (Dobbs et al. 2013). Others have argued that expansionary measures supporting financial asset prices are especially beneficial for the rich savers holding those assets (Financial Times 2014). These perspectives tend to focus on the impact of monetary policy on financial assets. They correctly highlight that a given change in their prices will produce different effects on households depending on the composition of their financial portfolios. However, they do not take into account that monetary policy also produces an impact on many other economic variables, including wages and incomes. Their implications are also heterogeneous across households – for example, depending on their employment status.

The distributional effects of monetary policy through unemployment

Recent evidence on the effects of monetary policy on unemployment in the euro area suggests that this is a quantitatively important source of heterogeneity (Lenza and Slacalek 2018). Such evidence is based on a two-stage approach. First, the macroeconomic outcomes of the ECB’s Assets Purchase Programmes are estimated for aggregate variables in France, Germany, Italy, and Spain. In a second stage, such effects are distributed across the individual households surveyed in the data from the Eurosystem Household Finance and Consumption Survey (HFCS 2018), using the information on their income composition and unemployment status.

Figure 1 shows how income across the distribution increases one year after an Assets Purchase Programme intervention via two channels:

1. transitions from unemployment to employment, and
2. increases in wages of all workers.

Figure 1 Decomposition of the total effect on mean income into the extensive (drop in unemployment) and the intensive margin (increase in wages)

Source: Eurosystem Household Finance and Consumption Survey, wave 2.
Note: The chart shows the percentage change in mean income across income quintiles in the euro area 4 quarters after the impact of the QE shock. It also shows the decomposition of the change into the extensive margin (transition from unemployment to employment) and the intensive margin (increase in wage). The numbers in brackets show initial levels of mean gross household income in each quintile. The statistics cover the euro area, which is here modelled as an aggregate of France, Germany, Italy, and Spain.

The figure illustrates two results. First, households in the lowest income quintile benefit substantially more from the monetary expansion – their income increases by roughly 3.5% on average, while for the remaining households it rises by about 0.5%. These changes result in a decline in the Gini coefficient for (gross) income from 43.1 to 42.9. Second, the bulk of the changes in income are driven by the effects of monetary policy on employment – the newly employed workers start earning wages rather than receiving unemployment benefits and their incomes increase. This share of the increase in income is decreasing across the distribution, from more than 90% in quintile one to about 40% in quintile five.

Direct and indirect effects of monetary policy

It is therefore important to consider all transmission mechanisms, including the effects on asset prices and on incomes, for a complete assessment of the impact of monetary policy on inequality. In a recent paper, we provide a simple and transparent quantification of these effects for the euro area, focusing on household consumption (Ampudia et al. 2018). Our theoretical reference framework is the work by Auclert (2017) and Kaplan et al. (2018), who analyse the transmission mechanism of monetary policy in the presence of household heterogeneity. These studies argue that the effects of monetary policy can be grouped into two broad categories: direct and indirect.

Direct effects are the immediate, partial-equilibrium consequences of the change in interest rates on households, holding their employment status and all prices and wages fixed. Direct effects include the impact of the different paths for interest rates on households’ net financial income (‘net interest rate exposure’). This is the channel highlighted by many commentators. It is heterogeneous across households depending on the composition of their asset and liability portfolios. For example, a reduction in policy rates will decrease interest payments for households with outstanding debt, especially if their loans are at a variable interest rate; it will also reduce the financial income of households that are not indebted, but hold short-maturity assets whose real return will temporarily fall. A second direct effect of monetary policy is to change households’ saving incentives (‘intertemporal substitution’). This effect is also heterogeneous, since it mostly operates on households that have a stock of liquid savings and are therefore able to temporarily adjust it without paying large transaction costs.

The indirect effect operates through the general equilibrium responses of prices and wages, hence of labour income and employment. After a reduction in policy rates, the direct increase in household expenditure (and firms’ investment) will lead to an increase in output and exert upward pressure on employment and wages. The additional increases in aggregate demand induced by higher employment and wages are the essence of the indirect effect. The indirect effect will also produce heterogeneous consequences to the extent that different sources of earnings – for example, employee income versus income from private businesses – or different pools of unemployed workers – for example, low versus high skilled – display different elasticities to the change in aggregate expenditures.

Quantifying heterogeneity in direct and indirect effects in the euro area

We provide an estimate of the aforementioned direct and indirect effects for three broad groups of euro area households, which are important from the perspective of the transmission of monetary policy. To define such groups, we build on empirical results on household heterogeneity in marginal propensity to consume out of transitory income shocks. Kaplan et al. (2014) argue that a key dimension of heterogeneity is the amount of liquid assets (such as cash or deposits) households hold. Households with adequate liquid assets are able to smooth transitory shocks to their consumption, while the spending of households with little liquid assets (‘hand-to-mouth’ households) is highly sensitive to shocks – thus these households tend to have a large marginal propensity to consume. An additional dimension of heterogeneity which we take into account is that hand-to-mouth households can also be ‘wealthy’; that is, they can hold a sizable amount of wealth in the form of an illiquid asset, such as their home. Following the classification method in Kaplan et al. (2014) and relying on information from the Household Finance and Consumption Survey, we conclude that roughly 11% of euro area households have close to zero liquid or illiquid assets (‘poor hand-to-mouth’), 12% of households are wealthy but own few liquid assets (‘wealthy hand-to-mouth’) and 77% have significant liquid assets.

To compute the size of direct and indirect effects of monetary policy for these three groups of households, we would ideally need household level information at the quarterly frequency. Due to the unavailability of such data, our results are based on a simple methodology which combines three ingredients. We first use estimates from Lenza and Slacalek (2018) on the dynamic effects of monetary policy shocks on interest rates and wage income. We then combines these estimates with detailed information on the composition of households’ assets and liabilities, available at lower frequency from the HFCS, to determine their hand-to-mouth status and their direct exposure to interest rate risk. Finally, we rely on existing estimates of the elasticity of substitution to interest rate changes and on estimates of the marginal propensity to consume of different groups of households to assess the overall importance of direct and indirect effects on aggregate consumption. In our exercise, we assume that the effects of monetary policy shocks are linear. Recent evidence suggests that consumers tend to respond more strongly to negative than to positive transitory income shocks (Christelis et al. forthcoming).

Figure 2 Decomposition of the effects of 100-basis-point cut in the policy rate on consumption of various households

Source: Eurosystem Household Finance and Consumption Survey, wave 2.
Note: The chart shows a decomposition of the effects of an initial 100-basis-point cut to interest rates on consumption in percent. The total effect consists of three parts, the effect of the net interest rate exposure, intertemporal substitution and income; sizes of these effects vary depending on households’ hand-to-mouth status. The numbers in the brackets below the bars show the fractions of households by their hand-to-mouth status in the euro area (which is here modelled as an aggregate of France, Germany, Italy, and Spain).

We report the following key results, which highlight that the direct effects of a cut in monetary policy rates differ between hand-to-mouth and other households. First, hand-to-mouth households are either entirely unaffected or negatively exposed to the direct, net interest rate exposure channel (dark blue area in Figure 2). They experience a gain in net financial income after a monetary policy easing because they are indebted and often have mortgages at a variable interest rate. They therefore benefit from a reduction in mortgage repayments. By contrast, the other households suffer an income loss after the monetary easing, because they have positive net assets. Thus the net interest rate exposure channel can indeed ‘hurt savers’; that is, households owning significant liquid assets. Our second results is that all households experience an increase in consumption once the intertemporal substitution channel is taken into account (light blue area in Figure 2). This is the consequence of a (temporary) reduction in savings induced by the lower real returns in the economy. This channel accounts for the bulk of direct changes in consumption for non-hand-to-mouth households, but it is not present for hand-to-mouth households, which do not accumulate liquid savings. The third result emerging from the figure is that indirect effects are beneficial for all households because everyone benefits from the increase in wage income following a monetary policy easing (orange area in Figure 2). In terms of the overall outcome, indirect effects are quantitatively more important – a result which is in line with those of recent analyses (Kaplan et al. 2018, Cloyne et al. forthcoming). This indirect effect is particularly strong for hand-to-mouth households, for whom it accounts for roughly 80% of the total effect on income, both because the employment effect is skewed toward these households and because they have a higher marginal propensity to consume. However, all households, even those experiencing a loss of financial income, gain from the increase in wage income and therefore increase their expenditure after the policy easing. The indirect income channel is thus a substantial driver of changes in consumption at the aggregate level.

Conclusions

Overall, our work finds that low short rates do hurt ‘savers’ – households owning significant liquid assets, via a direct effect – that is, via the reduction in their income from those assets. Low short rates, however, also benefit savers, like all other households, via an indirect effect – that is, the reduction in their unemployment rate and the increase in their labor income. The indirect effect is more important from a quantitative perspective. Since it is especially beneficial for hand-to-mouth households, it implies that expansionary monetary policy in the euro area led to a reduction in consumption inequality. We have also reviewed independent evidence suggesting that the ECB’s quantitative easing reduced income inequality, mainly through a reduction of the unemployment rate of poorer households. It should however be considered that the contribution of monetary policy to changes in inequality is small, compared to that of fiscal policy (e.g. the degree of redistribution of the tax scheme) or structural economic factors.