Did Quantitative Easing Increase Income Inequality?

Gerald Epstein

Click below to read full text.

Alejandro Reuss: This is Alejandro Reuss, co-editor of Triple Crisis Blog and Dollars & Sense magazine. We have on the phone today Gerald Epstein, professor of economics at the University of Massachusetts-Amherst and co-director of PERI, the Political Economy Research Institute. Welcome, Jerry.

Gerald Epstein: Hello.

Alejandro Reuss: Well, thanks for joining us. You have a recent PERI working paper out, co-authored with Juan Antonio Montecino, titled “Did Quantitative Easing Increase Income Inequality?” To start, could you tell us: what is quantitative easing exactly, and why was it one of the policy responses in the United States to the Great Recession?

Gerald Epstein: Quantitative easing is a monetary policy that was designed both by the Bank of England and the Federal Reserve to try to expand the credit available in the economy in order to fight the Great Recession of 2007-2008.

In some ways it’s like typical open market operations the central banks have often been doing. That is, it involves giving bank reserves to banks and using those reserves, or printing money, and using the reserves or the money to buy financial assets. In this case, unlike typical monetary policy—open market operations—the Federal Reserve initially used that money to buy asset-backed securities—mortgage-backed securities—that were key to the financial crisis itself.

The reason the Fed did this was because, by the end of 2008, after Lehman Brothers collapsed and the financial system almost totally melted down, the short term interest rate, the policy rate, the federal funds rate I guess they call it, it was down all the way to zero, and the Federal Reserve knew that they had to do more so they had to come up with some new ways of trying to inject credit into the system and get some bad assets off of the balance sheets of the banks to try to get them to lend more money.

Alejandro Reuss: Your paper focuses on what the effects of quantitative easing have been on income distribution. What are the hypotheses on either side about that question?

Gerald Epstein: Yeah, so this has become a big political fight—a big debate among economists as well—and there are several different positions. If you think about the channels of quantitative easing, buying these assets that the Federal Reserve did, you can see right away that there are different countervailing impacts.

On the one hand, when the Federal Reserve buys financial assets, by increasing the demand for them that will raise the prices of these assets and, as is well known, it’s the wealthy people, the richest people, who own the lion’s share of financial assets—stocks, bonds, government securities, etc. So when their prices go up, then, at least on paper, it’s the wealthiest people who gain, and when they sell these assets and they actually realize those gains. So that’s one channel, the asset-price effect, which is quite strong.

A second effect is that as interest rates are driven down and held down, then people who have savings and put them in the bank, or put them in other kinds of investments will find that the interest rate, the returns on those savings, have fallen. And since those tend to be primarily middle-class people, and especially who can’t find lucrative investments, and working class people who stick their money in the bank, savings account, etc.—they find that their interest rates have fallen and so they lose out, and this may be dis-equalizing as well.

On the other side though, lower interest rates and more credit, at least the Fed hoped, would lead to more employment, and more investment and this would generate, they hoped, more jobs and this would help people in the middle and the working class people at the lower end of the income distribution.

And, finally, the expectation was that, by putting more credit into the system, buying some more of these mortgage-backed securities and so forth, would make it easier and cheaper for people to refinance their houses that had gotten into so much trouble during the collapse of the housing bubble. And again, this was expected to help middle-class and working-class people.

So, on the one hand you have these first two effects, the asset-price appreciation and stock market appreciation and so forth, helping rich people, and on the other hand, you have more employment and more refinancing possibly helping working-class and middle-class people. So it really came down to a kind of empirical question about the overall impacts.

Alejandro Reuss: So just to follow up on that last point, you talked about the controversies among economists on these questions. Is it the case that there was agreement about what these channels were, and then it becomes the empirical question of determining what the magnitude of each of these different effects is?

Gerald Epstein: Well, if we think about the economists’ debate on this and analysis, I think there’s pretty much an agreement about what the channels are, but of course it spilled over, you know, into the political and policy realm where things were not so carefully delineated. So on the one hand you have this interesting mix of left-wing populists and right-wing populists who have argued that the Federal Reserve and its policy has helped the banks and created more inequality, and there is not always such a clear delineation of these channels. And then on the other hand you have other economists and Federal Reserve officials just saying, well you know, clearly monetary policy has prevented the collapse of the economy, so clearly this was necessary and it helped everybody. But when you get down to kind of a careful debate among economists, these channels are pretty much agreed to, and then it comes down to an empirical question.

Alejandro Reuss: Alright, so what were the findings that you report in the recent paper about the distributional impacts of quantitative easing?

Gerald Epstein: Well, Juan and I were the first people to really look at this empirically. Josh Bivens, from the Economic Policy Institute, wrote an interesting paper where he kind of amassed different estimates that others had put together that would bear on the issue, and his result suggested that quantitative easing, by generating more employment, reduced inequality. But what we did was actually look at the Survey of Consumer Finances data put together by the Federal Reserve which has detailed information on the sources of household income throughout the entire income distribution, and so we looked at these channels, of the three rounds of quantitative easing from 2008 to 2013.

What we initially found was that over this period, there was a big increase in inequality, especially measured by the so-called 99 to 10 ratio, that is, if you look at the 99th percentile—that is, the 1% richest people—versus the 10% of poorest individuals, that there was a big increase in inequality over this period, somewhere between 8 and 12% increase in this 99-10 ratio. And, by looking at these channels I described before, we found that in fact these channels were very dis-equalizing. In fact, they more than explained the 8-12% increase in the 99-10 ratio, so that there were other things going on in the economy not due just to these channels that actually equalized things.

So if you just look at these four channels, they were extremely dis-equalizing and the main aspects of that were the equity price increases, stock market increases, really gave a lot of income to the richest and since poor people and middle-class people don’t hold much stock it didn’t help them. Plus, the decrease in interest rates hurt working-class and middle-class people more than wealthy people, because wealthy people had access to other kinds of investments. Employment did increase, to some extent, and that did help workers, but during this period real wages actually fell, so if you look at the total impact for labor income from employment that actually went down during this period. And the mortgage refinancing channel was very interesting because, what we found was that whereas the wealthiest people and middle-class people were able, to some extent, to refinance their houses at lower cost, working-class and poor people—they were blocked. They weren’t able to refinance because their houses were largely underwater, their credit ratings had gone down, so they weren’t really able to benefit from this.

Now, though these channels were very dis-equalizing, we can’t attribute all of these impacts to quantitative easing. There was a lot going on during this period, so what we then did was look at—well, how much of this is really due to quantitative easing, and how much of this was due to other factors? So to study that we looked at a lot of work that had been done by the Federal Reserve and other economists that tried to estimate the impacts of quantitative easing on stock-price appreciation, on employment generation, and so forth.

And we took a very conservative set of estimates from these studies, which said anywhere from 5-10% of the stock price appreciation was due to quantitative easing and about 1.1 percentage points decrease in unemployment was due to quantitative easing. In putting those numbers together, what we find is that quantitative easing was probably due to about 13% of the total increase in inequality during this period—that is, increases in the 99-10 ratio. So, quantitative easing was dis-equalizing and maybe accounts for 13-15% of this dis-equalizing effect, but this is a very conservative estimate because we didn’t want to overstate it.

Now what’s interesting about this result is that it suggests, I think, a real paradox in our understanding of the relationship between monetary policy and inequality. I mean, typically what economists and political economists have argued is that when the Federal Reserve tightens policy, that is when it increases interest rates, that leads to more inequality. But what these results suggest is that, well, when the Federal Reserve this time around lowered interest rates and did this quantitative easing, that led to more inequality.

So how can both of these things be true at once? What we argue is that the main problem is the way in which the housing bubble and the crisis led to massive problems for households—the value of their houses, the fact that their houses were underwater—they couldn’t benefit from these refinancing programs or from quantitative easing, on the one hand. And the other thing is that the weak power of labor due to years of busting labor unions, labor-law changes that worked against unions, globalization and so forth, that workers weren’t really able to benefit that much from the employment increases that did occur largely because their wages were stagnant.

So I guess the end result that we come up to is that Federal Reserve policy can make marginal differences one way or the other, but in order to really reduce inequality, much bigger changes are needed to bulk up the power of unions to organize, bigger programs to help underwater homeowners, and more public investment is probably what’s needed to really turn this thing around.

Alejandro Reuss: That last point speaks to the question of the policy conclusions that you see as following from this research. Can you talk a little bit about that, both in regard to the conduct of counter-cyclical macroeconomic policy in the context of a serious downturn, and those broader policy questions that you raised?

Gerald Epstein: Sure. So, there’s first of all a short term issue on policy that even though the decrease in interest rates and quantitative easing increased inequality, that doesn’t suggest that when the Federal Reserve starts raising interest rates that that’s going to help inequality. On the contrary, I think the typical pattern is that increases in interest rates hurt workers and I don’t see anything that changed that’s going to reverse that. So, when Janet Yellen and the Federal Reserve starts raising interest rates that’s certainly not going to reduce inequality; I think quite the contrary.

More generally, though, if the Federal Reserve, in conducting the counter-cyclical monetary policy, particularly in a serious downturn like we had, wants to really help workers and really wants to help debtors that got hurt so badly by the housing bubble, they’re going to have to engage in much more activist policies. Rather than buying up financial assets that are held mostly by the wealthy, they have to get credit directly to those who need the credit the most, that is, states and locales, underwater homeowners, directly into public investment to increase infrastructure, jobs, etc. So the Federal Reserve is going to have to devise much more direct mechanisms of lending money directly rather than going through the financial markets, which, in the way that they did it just raises the value of securities owned by wealthy people and very little of that trickles down to those in the middle or those at the bottom, or as needed public investment.

Alejandro Reuss: Well, very good. Thank you very much Jerry, for joining us here on Triple Crisis today.

Gerald Epstein: Well thanks very much for having me.

Triple Crisis welcomes your comments. Please share your thoughts below.

Triple Crisis is published by