A Mystery in Fixed Income

Paul Krugman, "How Negative Can Rates Go?," March 3, 2015, The Conscience Of A Liberal, https://archive.nytimes.com/krugman.blogs.nytimes.com/2015/03/06/the-below-zero-lower-bound/

We now know that interest rates can, in fact, go negative; those of us who dismissed the possibility by saying that people could simply hold currency were clearly too casual about it. But how low? Evan Soltas is getting some attention with an interesting attempt to find the true lower bound on interest rates; David Keohane cited a similar, more detailed estimate a month ago.

But it seems to me that all such estimates that I’ve seen involve a misconception. (Maybe it’s me, so I’m happy to stand corrected.)

In both the Soltas and Keohane pieces, there’s a discussion of storage costs for currency, which are not as trivial as one might have assumed. But they go on to say that the real lower bound comes from the fact that bank deposits are more useful than currency in a safe, because you can write checks and all that. Basically, in the modern world deposits are actually more liquid than cash, at least for most transactions that don’t involve controlled substances or concrete overshoes.

But you need to think about the incentives for holding a dollar in one form or the other at the margin, and I think that changes the story.

In normal times, we invoke the convenience of money — its extra liquidity — to explain why people hold money at zero or at any rate low interest rates when there are other safe assets offering higher yields. We think of money demand as determined by people increasing their holdings up to the point where the opportunity cost of holding money, the interest rate on other safe assets, equals its utility from increased liquidity.

Once interest rates on safe assets are zero or lower, however, liquidity has no opportunity cost; people will saturate themselves with it. That’s why we call it a liquidity trap! And what this means is that the marginal dollar of money holdings is being held solely as a store of value — the medium of exchange utility is irrelevant.

This in turn should mean that the usefulness of deposits is irrelevant in trying to find the true lower bound. The marginal holder is simply looking for a store of value, and the only question should therefore be storage costs.

And I am pinching myself at the realization that this seemingly whimsical and arcane discussion is turning out to have real policy significance.

Paul Krugman, “Quantitative Easing and Monetary Aggregates,”The Conscience Of A Liberal, February 26, 2015, https://archive.nytimes.com/krugman.blogs.nytimes.com/2015/02/26/quantitative-easing-and-monetary-aggregates/

Quantitative Easing and Monetary Aggregates

I haven’t had time for the broader discussion of cognitive closure, but one more thing about Meltzer and all that: I get especially annoyed when economists who have been wrongly predicting inflation say that it’s not their fault — who could have known that banks would just sit on all those reserves? The answer is, anyone who had paid attention should have known that would happen.

Let me quote myself, from my 1998 — yes, 1998 — paper on the liquidity trap: “The point is important and bears repeating: under liquidity trap conditions, the normal expectation is that an increase in high-powered money will have little effect on broad aggregates, and may even lead to a decline in bank deposits and a larger decline in bank credit.”

I presented data from the 1930s that seemed to confirm that point; a few years later Japan gave us another experiment, when it tried quantitative easing. Here’s how it went, with the monetary base and M2 both shown with January 2001=100:

So theory and experience both predicted exactly the sterility of monetary base expansion that we saw in practice. And, you know, that’s the kind of successful prediction that is supposed to change peoples’ minds: if you’re that wrong about how an experiment turned out, and someone else made a prediction you considered foolish but turned out completely right, you’re supposed to concede that just maybe, possibly, they were on to something.

The fact that essentially nobody on that side of the debate has budged in the slightest tells us that whatever it is they’ve been doing, it’s not scientific research.

"...Expansionary monetary policy has heterogeneous effects on the labor force, with labor market tightness playing an important mediating role. We show empirically that expansionary monetary policy benefits the employment of workers with weak labor force attachment more in tight labor markets than in slack ones. This pattern holds across racial, education, and sex categories, as the employment benefits for Blacks, high school dropouts, and women increase with labor market tightness. The beneficial impact of monetary policy on less-attached workers is economically sizeable and long lasting. Using a New Keynesian model with workers of heterogeneous types, we analyze how labor market tightness transmits changes in monetary policy into employment growth of workers of different types. The model predicts that the expansionary effect of monetary policy on the employment of less-attached workers is stronger in tighter labor markets. We further show that a monetary policy that follows an average inflation targeting rule particularly benefits less-attached workers. By keeping rates low for longer, employment becomes more inclusive. Similarly, a flatter Philips curve enables the central bank to maintain low rates, implying that expansionary monetary shocks lead to larger and more persistent increases in the employment of low labor force participation workers. Our empirical and theoretical results both suggest that sustained expansionary monetary policy, which tightens labor markets, facilitates robust employment growth among less attached workers. Our findings thus imply that the Federal Reserve’s recent change in its conduct of monetary policy from strict to average inflation targeting will benefit the employment of female, minority, and low skilled workers. At the same time, expansionary monetary policy increases inflationary pressure and may also foster wealth inequality by raising asset prices (Amberg et al., 2021; Peydró et al., 2021). Managing the tradeoff between broad-based employment goals, inflation targets, and wealth inequality is an important topic of further research...."

Nittai Bergman, David Matsa, and Michael Weber," Inclusive Monetary Policy: How Tight Labor Markets Facilitate Broad-Based Employment Growth,"Becker Friedman Institute, January 2022, https://bfi.uchicago.edu/wp-content/uploads/2022/01/BFI_WP_2022-03.pdf

“…The employment response of Whites, however, differs from that of Blacks. Column 2 of Table 3 reports estimates of equation (3) for Whites. In contrast to Blacks, the b1 coefficient for Whites is much smaller and not statistically significant...The difference in the Black and White coefficient estimates is highly statistically significant (p < 0.01)...."

“..Panel B of Table 3 presents a similar heterogeneity analysis with respect to educational attainment, reporting results for those who did not complete high school in column 3, high school graduates in column 4, those with some college education in column 5, and bachelor’s degree holders in column 6.8 We find that in response to monetary easing, the increase in employment growth among workers who did not complete high school is larger when labor markets are tight than when they are slack (column 3). The b1 coefficient implies that a one standard deviation drop in the federal funds rate is associated with 0.39 percentage point greater two-year employment growth in tight labor markets (90th percentile) than in slack ones (10th percentile). This magnitude corresponds to approximately 18% of unskilled workers’ mean two-year employment growth...For workers with greater educational attainment, in contrast, the b1 coefficient estimates are close to zero and not statistically significant (columns 4-6)…”

“…Our empirical analysis explores monetary policy’s heterogeneous effects with respect to workers’ race, education, and sex. We investigate how expansionary monetary policy promotes employment growth for each group across local labor markets with different tightness. We find that for demographic groups with lower average labor market attachment—Blacks, the least educated, and women—expansionary monetary policy has a larger effect on employment growth in tighter labor markets. Because expansionary monetary policy tightens labor markets...this finding implies that sustaining expansionary monetary policy over longer time periods is particularly helpful to these demographic groups...Our results show that for demographic groups with low average labor market attachment—Blacks, the least educated, and women—monetary expansions have a larger effect on employment growth in tight labor markets, which we measure using the market’s aggregate prime-age employment-to-population ratio. This effect is economically large. For example, we find that a one standard deviation drop in the federal funds rate increases subsequent two-year Black employment growth by 0.91 percentage points more in tight labor markets (90th percentile) than in slack labor markets (10th percentile).Similarly, for workers who did not complete high school, a one standard deviation drop in the federal funds rate increases employment growth over the subsequent two years by 0.39 percentage points more in tight labor markets than in slack ones. This additional impact of monetary policy in tight labor markets is sizable, corresponding to 9% and 18% of the mean employment growth rates for Blacks and high school non-completers over the sample period, respectively....

The effect is seemingly persistent

“..The effects on less-attached workers are persistent. We find that monetary policy’s incremental effect on less-attached workers’ employment growth in tight labor markets peaks 7 to 9 quarters after interest rates decreases. Although monetary policy’s incremental effect wanes over time, its cumulative effect is long lasting. For example, the differential effect of monetary policy on cumulative Black employment growth in tight versus slack labor markets persists even four years after the federal funds rate decreases…”

The Evidence

“…Table 3 presents OLS estimates of equation (3). Each column in Table 3 examines the employment growth of a different demographic group. Panel A of the table examines heterogeneity with respect to workers’ race, presenting results for Blacks in column 1 and Whites in column 2. For Blacks, the coefficient on the interaction between the federal funds rate and local labor market tightness, b1, is negative, sizable, and statistically significant. This coefficient implies that a monetary easing is associated with greater Black employment growth in tight labor markets as compared to in slack ones.To assess the magnitude of this estimate, consider the effect of a one standard deviation (2.25 percentage point) decrease in the federal funds rate. Our estimate implies that, over the subsequent two years, this drop in the federal funds rate is associated with a 0.91 percentage point larger increase in Black employment growth in labor markets at the 90th percentile of employment-to-population (86%) than in labor markets at the 10th percentile of employment-to-population (49%).This additional boost in employment growth in tighter labor markets is sizable, corresponding to 9% of the mean two-year Black employment growth over the sample period…”

"...By buying Treasurys and mortgage-backed securities, the Fed acquired or offset some 45% of all federal debt issued during the Obama era—about four times the share of federal debt the Fed purchased during World War II..... Historically, banks held few excess reserves as the Fed did not pay interest on them. The money supply changed when the Fed altered bank reserves through lending or buying and selling securities. Now if market interest rates rise and the Fed does not act by raising the rate it pays on reserves, the money supply increases as banks increase lending. As a result, to maintain any given money supply, the Fed must respond to changes in market interest rates. In doing so the Fed becomes an interest-rate follower, not a leader...."
Phil Gramm and Thomas Savings, "The Fed’s Obama-Era Hangover,"Wall Street Journal, January 1, 2018, https://www.wsj.com/articles/the-feds-obama-era-hangover-11546374393