Inequality has long been a foremost public concern, but not so for central banks. While the orthodoxy according to which central banks engage in “neutral” policy-making is slowly crippling away, we’re still enormously far from taking the distributional consequences of money creation and allocation seriously. Intensifying the debate on the inequality effects of the central bank’s monetary policy is a small step in that direction. This blog suggests several starting points for that.
Why does inequality matter in regards to monetary policy?
The below presents 8 main lessons distilled from an academic literature review on the nexus of inequality and monetary policy. Statements on the inequality effects of monetary policy instruments synthesise the findings of around 40 empirical papers. A summary of these papers can be requested from the author.
1. Monetary policy is always distributional.
Monetary policy always raises or lowers the income and wealth of some agents more than of others. This is because it works by affecting the cost of credit in the economy which has different implications for different people. For example, asset-poor households suffer from higher borrowing costs while wealthier households who often lend money to others benefit from a higher price of credit. Also, some are simply more sensitive to a change in the cost of credit than others. For example, firms with higher leverage ratios respond more strongly than firms with a lower leverage ratio. Given the heterogeneity in households’ and firms’ financial characteristics, monetary policy will always give/take more to/from some than others.
2. Distributional effects can be long-lasting.
The distributional effects of monetary policy can be long-lasting, outliving the duration of business cycles. This is, for example, because monetary policy shocks can cause a recession which can condemn people into unemployment (see Blanchard, 2018; Yagan et al., 2019, BIS, 2022) and permanently reduce their life-time earnings (J. Schmieder et al., 2021; Lachowska et al., 2019; Quintini & Venn, 2013). The inequality induced hereby often persists (BIS, 2022). Even if the central bank lowers interest rates as often it raises them, its effects would not balance out. Studies suggest that higher interest rates lower the incomes of poor people more than lower interest rates raise their incomes, which means that, over time, monetary policy hurts the poor (Furceri et al. 2018; Guerello et al., 2018). Far from being neutral, the cumulative effect of monetary policy on the entire income distribution might be regressive (Kronick & Villarreal, 2020).
3. If the central bank raises interest rates, this hurts low-income households. In reverse, lowering rates tends to raise their incomes.
If the central bank raises interest rates, banks will experience higher refinancing costs, which they tend to pass onto households and firms. Households will then face higher interest payments, for example on their mortgage. Firms might respond to higher credit costs by not implementing planned wage increases or by firing workers. This particularly slashes the salary incomes of those who already don’t earn much (see Guerello et al., 2018; Mumtaz et al., 2017; Coibon et al., 2017; Gornemann et al., 2016; Park et al., 2021; Holm et al., 2021; Aye et al., 2019). Why? Because low-paid workers are often the first who get fired (Dossche & Hartwig, 2019; Slacalek et al., 2021).
In reverse, a lowering of the interest rate is mostly argued to benefit low-income households: cheaper credit reduces debt-servicing costs and stimulates hiring by firms as well as rising wages. For the income situation of low-income households, the effect flowing from firms’ decisions on hiring and wages is more important than changes in debt-servicing costs (Samarina & Nguyen, 2019; Corrado & Fantozzi, 2020). Here again, the effect that matters most for labour earnings (and income overall) is the employment gain for those who were previously unemployed (Lenza & Slacalek, 2018; Casiraghi et al., 2018; Park, 2021, Lee, 2021; Amberg et al., 2021).
4. High-income households also benefit from falling interest rates. But they are hurt less by interest rate increases.
Also the wages of high-income households are increasing when the central bank lowers the interest rate. Although they lose out as creditors from an interest rate decrease, they win out as debtors. High-income households tend to have large outstanding loans, and empirical evidence suggests that the positive effect of decreasing debt-servicing costs outweighs the negative effect of lower interest income for high-income households (Andersen et al., 2022). Other channels add to their income gains: business income and capital income (from rents and stock returns) increase after an expansionary monetary policy shock. As high-income households tend to be more often self-employed, and even more so, own financial assets, compared to lower-income households, the income gains through these channels largely accrue to them. This is clearest for unconventional expansionary monetary policy (Saiki & Frost, 2014; Montecino & Epstein, 2015, Mumtaz & Theophilopoulou, 2017; Taghizadeh-Hesary et al., 2018; Davtyan, 2018; Hafemann et al., 2018, Saiki & Frost, 2020) but also seems to be the case for conventional policy (Andersen et al. 2022).
A stark interest rate hike, as mentioned before, leads to lower salaries, including for high earners. Salaries are particularly sensitive to an interest rate hike-induced economic downturn. But high-income individuals are less dependent on salaries. More so than others, they also get income from financial sources. For example, they receive interest income on their savings – a type of income that even increases with an interest rate hike. Hence, high-income individuals are hurt less by interest rate increases than low-income households
5. Expansionary monetary policy has boosted the wealth of the already wealthy.
The value of assets tends to rise when a central bank lowers its interest rates or purchases assets on the financial market. A standard rate decrease by the central bank directly affects short-term interest rates; for example, it decreases the interest rate on deposit and savings accounts. This tends to incentivize investors to move their money into riskier assets such as stocks that pay higher returns. Stock prices then increase. Interest rate changes also have an indirect effect on long-term interest rates. Bond prices increase because all potential new bonds that are issued after the interest rate decline will provide less yields than the existing ones, making existing bonds increase in value. Asset purchases by the central bank directly affect long-term interest rates because central banks buy long-maturity assets such as government bonds. As a consequence of the additional demand, their price increases, yields decrease, and all of this again incentivizes a portfolio reshifting towards riskier assets such as stocks. What about real estate? Housing assets are indirectly affected by monetary policy in that the central bank lowering its rate for banks also eventually brings down mortgage rates, increasing the demand for housing and pushing up the price. Hence, in general, asset prices increase in response to expansionary monetary policy.
Asset price increases make the rich even richer, because it is mostly them who own financial assets. In some jurisdictions such as the USA, Spain, Italy, and Portugal, real estate is more evenly distributed and the most important wealth factor for the middle-class. House price increases can then act as a countervailing force to the inequality-creating effects of stock and bond price increases (O’Farrell et al., 2016). However, even if that is the case, the wealth increase due to stocks far outweighs the valuation gains in real estate (Domanski et al., 2016; Adam & Tzamourani, 2016; Andersen et al., 2021). Hence, expansionary monetary policy has, in general, increased wealth inequality.
6. Interest rate hikes aggravate existing inequalities.
Inequalities exist along various dimensions and often come with cases of one group having power over another. For example, in Western societies, and as a matter of empirical tendency over the last decades, young people tend to have less power than older people; highly-educated people reflect a stronger position in society than lower-educated people; native-born individuals are more fortunate than migrants; precariously employed workers are worse off than most of their co-workers; and, capitalists have more market and bargaining power than workers (Guschanski & Onaran, 2021; Sokolova & Sorensen, 2021).
Raising the interest rate deepens inequality between these groups by worsening the position of the already subordinate group. For example, the central bank’s interest-rate hammer primarily comes down on working people. If firms face higher loan repayment costs after the central bank raised interest rates, they don’t *have to* balance out these costs by paying their workers less or firing them. They could also cut costs in other areas or tolerate less profits. But, empirical evidence shows that, overall, firms tend to spend less on staff as a response to interest rate hikes. Hence, relative to what accrues to the owners of capital, working people lose out.
Furthermore, young people, little-educated workers, and low-wage earners are quite sensitive to cyclical fluctuations in terms of the risk of getting fired (ECB, 2021, ECB, 2022). Hence, they would suffer first if an interest rate increase brought about an economic downturn. Similarly, those working in precarious employment conditions (e.g., temporary contract workers, agency workers, part-time workers, mostly female) are more likely to lose their jobs when firms cut costs in response to interest rate hikes.
7. Inflation-centric monetary policy hurts workers
Inflation-centric monetary policy favours the wealth-rich and retirees. Why? If a central bank prioritises stable inflation over stable employment, average unemployment ends up being higher and more volatile (see Gornemann et al., 2021). In reality, this means that inflation targeting makes people unemployed who would have had a job under a more employment-focused monetary policy regime. Job losses don’t matter to retirees and matter little to the wealth-rich, hence, they win out.
8. Monetary policy gets transmitted less effectively when inequality is high.
Monetary policy gets mediated through the income and wealth distribution and hence can be more or less effective depending on the specifics of the prevailing distribution (see Auclert, 2019). How does this happen exactly? Consider that lower-income households tend to spend a higher portion of their earnings on goods & services so that a policy measure that raises their income leads to more economic activity. Vice versa, income-rich people tend to save more which consequently doesn’t boost economic activity as much. In a highly unequal society where the already rich capture most of the benefits of added economic activity, an expansionary monetary policy generates more benefits for the rich. Due to their higher propensity to save rather than spend, this constraints the multiplier effect of monetary stimulus and makes it less effective. Furthermore, very poor individuals and very rich individuals are less sensitive to the cost of credit induced by interest rate changes. Very poor households might own nothing they can offer as collateral to get a loan or don’t even have a bank account, while wealthy households are too rich to care. Hence, in societies with lots of very poor and very rich people, monetary policy loses effectiveness.
How does monetary policy create inequality?
Through wealth (capital income)
“… by raising wealth of some more than others”
- Savings remuneration (and cost of debt) channel: the relevant heterogeneity dimension here is net wealth. An interest rate change affects the conditions of net borrowers and savers. An interest increase will bring about higher returns for savers but higher costs for borrowers, and hence distributes money from borrowers to savers.
- Portfolio composition channel: the main heterogeneity factor here is the portfolio, precisely the type of asset class in one’s portfolio; capital income might respond differently to monetary policy shocks for high-income and low-income households. This is because high-income households have a different asset portfolio than low-income households and monetary policy might raise the returns of one asset class more than of another.
- Financial segmentation channel: main distinguishing factor between households: access to financial markets; for example, higher income households could have a greater access to financial markets which could imply that they can reallocate their portfolios such as to benefit from a monetary policy shock
Through the labour market (labour income)
“… by raising labour income for some more than others”
- Earnings heterogeneity channels: it states that labour earnings might respond differently to monetary policy shocks for high-income and low-income households. For example, expansionary monetary policy could more strongly raise the labour income of low-income households, simply because it could reduce unemployment, which usually falls upon low-income groups, and hence raise their labour income more than the labour income of high-income, employed households (employment channel). However, it can also be argued that expansionary policy affects the labour income of high-income households more strongly as the wage of low-income households is usually stickier and as it could lead to firms substituting low-skill workers for high-skill workers (skill premium channel). Here, the distinguishing factor between households is the sensitivity of wages to employment changes.
Through different income compositions
“… by some individuals benefiting more from one type of income than from another”
Income composition channel: it states that different types of income might respond differently to monetary policy and as income groups differ in what type of income (e.g., capital income, labour income) constitutes their primary source of income, monetary policy can end up increasing/decreasing the disposable income of certain income groups more than that of others. Here the distinguishing factor between households is the share of capital vs labour income.