Lael Brainard on monetary policy and employment
Joe Weisenthal directed me to a recent speech by Lael Brainard. Here’s one excerpt:
The new framework calls for monetary policy to seek to eliminate shortfalls of employment from its maximum level, in contrast to the previous approach that called for policy to minimize deviations when employment is too high as well as too low. The new framework also defines the maximum level of employment as a broad-based and inclusive goal assessed through a wide range of indicators.
Some pundits regard this as a significant change, perhaps indicating that the Fed will stop being so mean to workers looking for jobs. While I don’t oppose the change, I don’t see much difference from previous policy.
This is a really complicated issue with lots of moving parts and lots of subtle distinctions. So here are some of the tools I will work with:
- The Plucking model. The business cycle is mostly (not entirely) shortfalls from full employment, not symmetrical deviations around a natural rate of unemployment.
- The Natural Rate Hypothesis predicts that (for positive inflation rates) employment does not depend on the average rate of inflation. In the long run, workers adjust to changes in the expected rate of inflation, and hence money is roughly super-neutral in the long run.
- The Fed erred in raising rates 9 times during 2015-18, as it relied too heavily on highly flawed Phillips Curve models of the economy. As a result, inflation has averaged less than 2% over the past decade.
- The Fed recently switched to average inflation targeting (AIT), which will cause inflation to average roughly 2% in the years ahead.
- Under AIT, there is no such thing as hawks and doves.
If you combine the first two points, then the implication is that a Fed policy that minimizes shortfalls in employment is basically the same as a Fed policy that minimizes deviations in employment. After all, the only long run equilibrium in the labor market is full employment, according to the natural rate hypothesis.
[Note: A permanent and unchanging 40% unemployment rate caused by a $35/hour minimum wage is “full employment” as defined by macroeconomists]
The existence of a 2% AIT means we’ve moved beyond hawks and doves; all future disputes will be about fluctuations in inflation and employment, not the average rate of inflation.
That doesn’t mean there is nothing new at the Fed:
For nearly four decades, monetary policy was guided by a strong presumption that accommodation should be reduced preemptively when the unemployment rate nears its normal rate in anticipation that high inflation would otherwise soon follow. But changes in economic relationships over the past decade have led trend inflation to run persistently somewhat below target and inflation to be relatively insensitive to resource utilization. With these changes, our new monetary policy framework recognizes that removing accommodation preemptively as headline unemployment reaches low levels in anticipation of inflationary pressures that may not materialize may result in an unwarranted loss of opportunity for many Americans. It may curtail progress for racial and ethnic groups that have faced systemic challenges in the labor force, which is particularly salient in light of recent research indicating that additional labor market tightening is especially beneficial for these groups when it occurs in already tight labor markets, compared with earlier in the labor market cycle.33 Instead, the shortfalls approach means that the labor market will be able to continue to improve absent high inflationary pressures or an unmooring of inflation expectations to the upside.
The phrase “changes in economic relationships” is a polite way of saying the Fed screwed up in 2015-18 and tightened too soon because they relied on flawed Phillips Curve “economic relationships”. They learned a lesson and promise not to do so again. But despite all the woke window dressing about jobs for minorities, this isn’t really about jobs at all, it’s about achieving their target. The mistake made during 2015-18 was a mistake whether you focus on employment shortfalls or employment deviations.
So what does the Fed’s dual mandate mean today? Roughly this:
We promise to deliver 2% PCE inflation on average, and allow modest fluctuations around that average only to the extent that those fluctuations reduce employment shortfalls. As a practical matter, reducing employment shortfalls means keeping employment close to the natural rate, which implies also reducing employment deviations.
AIT really is new—and an improvement. The increased awareness of the unreliability of Phillips Curve models really is new—and an improvement.
The talk about employment is just a bunch of pretty words. Not that there’s anything wrong with pretty words, if they help to provide political cover for AIT and moving on from Phillips Curve models.
Tags:
24. February 2021 at 22:06
Re the ‘Plucking model’, you say this is about the business cycle being mostly but not entirely about employment shortfalls. Is the ‘not entirely’ a reference to the possibility that if we ever do see persistently strong growth in NGDP, we could expect unemployment to fall below the natural rate? And hence the veracity of the Plucking model stems from the way monetary policy is conducted? For example, one could imagine a Fed that oscillated between mildly overheating the economy and soft landings, in which unemployment deviations were more symmetrical.
24. February 2021 at 22:52
Rajat, It’s more than a possibility, strong NGDP growth does sometimes push employment slightly above the natural rate, as in 1968-69.
But the assumption of the model is that more volatility means a lower average level of employment, which sounds right to me.
25. February 2021 at 01:33
TIPS spreads are really starting to look like the market believes the Fed’s commitment to AIT.
I hope they deliver. And also that this means that the rest of economic policy can afford to be approximately classical, ie fiscal policy should be done by cost-benefit analysis and not with the excuse of ‘providing stimulus’.
25. February 2021 at 05:29
This is an interesting interpretation on the reason for the Fed’s failures 2008-20. Other explanations are that a) they really had an inflation ceiling target or b) they had a “how much QR can we get away with” constraint.
It’s still and interesting question what is the optimal inflation target. What is the trade off of nominal inflation speeding adjustment of prices subject to >zero rate of changes vs damaging expectation of future relative price relations.
25. February 2021 at 05:29
Scott,
I think the political cover argument makes sense, and could add value. He gets a lot of dumb questions during hearings, and having another policy to hang his hat on to fend off bad ideas could be very useful. Like it or not, he will continue to get Philips curve-ish questions.
I also wouldn’t rule out the possibility that this was added to reach consensus behind the scenes. Maybe this language assuages some concerns/objections of some faction in the Fed.
AIT strikes me as very welcome news.
So which organization is slower at adapting to crisis and new information, FRB or FDA?
25. February 2021 at 06:09
You should write opinion columns in the WSJ. One thing is clear, monetary policy is really not understood by the public. (Perhaps limiting that statement to MP is too optimistic). Or many commenters. The only economists that write columns have a political agenda. Or they just get some quotes from various traders etc. Maybe all news is damaged in this way——-and while it is clear to me that it is hard to track the trail that gets you to your conclusions, your conclusions at are coherent——-which is often good enough for the public and people like me. Perhaps because for whatever reasons I have always agreed with your prescriptions—-or at least their purpose.
Just like Brainard needed to throw in propaganda which was harmless ——sometimes propaganda is all we have. Propaganda is not “bad”—-Bad propaganda is “bad”.
25. February 2021 at 06:22
A few years ago, Andy Harless made some comments about “inflation hawks” http://blog.andyharless.com/2013/02/why-doves-are-really-hawks.html
That’ resonating now in the context of AIT because a modern “hawk” can be someone who will push back aggressively against anyone who freaks out over two quarters of inflation above 2%—or two quarters below.
The modern “hawk” is going to pay close attention to hourly wage trends and structural movements in labor force participation.
25. February 2021 at 07:05
I notice that average wages and ECI for private sector employees seems to track average RGDP very closely, as opposed to inflation, since the Great Moderation began, except for recessions of course. Growth in these compensation metrics are usually below NGDP growth, however, meaning average wage/NGDP decreases over time, whereas, in periods of relative stability, average wage/RGDP is on such a constant, gentle decline, it looks like a law of nature.
https://fred.stlouisfed.org/graph/?g=BnjA
Am I missing something, or should workers demand wage growth equal to NGDP growth, at least at full employment?
25. February 2021 at 08:14
re: “The Fed erred in raising rates 9 times during 2015-18”
That’s dead wrong. The FOMC eased in the summer of 2017, causing the next downswing (as predicted).
During the U.S. Golden Era in Capitalism (which was not optimized as there were 2 recessions, Aug 1957–April 1958 and Apr 1960–Feb 1961 ), the annual compounded rate of increase in our means-of-payment money supply was about 2 PER CENT. And during this same period, 1955-1964, the rate of inflation, based on the Consumer Price Index, increased at an annual rate of 1.4 PER CENT. Unemployment averaged 5.4 PER CENT.
The fact is that during the U.S. Golden Era in Capitalism private-sector debt dominated, while public sector debt, *real net Federal debt* remained absolutely unchanged. The economy was financed in 2/3 by an increase in savings that was invested (or by velocity). Today, almost all economic activity is diametrically opposed, i.e., financed by new money.
From 1954-1966, the 12-year annualized rate-of-change in R-gDp was 5.5% (the real output of goods and services).
Things ended in 1965. That’s when the commercial banks, DFIs, began to outbid the non-banks, NBFIs, for loan-funds (resulting in disintermediation of the thrifts). That’s when William McChesney Martin Jr. re-established stair-step case functioning (and cascading), interest rate pegs (like during WWII), thereby abandoned the FOMC’s net free, or net borrowed, reserve targeting position approach in favor of the Federal Funds “bracket racket”.
I.e., the Fed once again, began targeting interest rates and accommodating the banksters and their customers whenever the banks saw an advantage in expanding loans, thereby usurping the Fed’s “open market power” (in 1955 rhetoric, refilling the “punch-bowl”, at first coinciding with the rise in chronic monetary inflation and in anticipated inflation and stagflation, and eventually and inevitably, after the 1981 “time bomb”, secular strangulation).
25. February 2021 at 08:26
re Powell: ” ‘the Fed chairman said he didn’t think price gains would be “persistent.’ ”
You find that a lot of high brows won’t admit their mistakes. But Powell is indeed correct. We’re experiencing more of reflation rather than inflation (with the unique exception of housing).
25. February 2021 at 08:49
The current AIT framework still frequently leads to pro-cyclical monetary policy. I’m optimistic about near-term improvements from the positive policy change.
But what are the odds enough people become convinced of an NDGP or aggregate nominal income target in time to avert the shortcomings of AIT?
25. February 2021 at 08:53
*NGDP
25. February 2021 at 09:19
re: “convinced of an NDGP or aggregate nominal income target”
It’s not a new idea. In 1961: “A theoretical explanation was advanced by Dr. Lester Chandler (“A monetary expert”), Ph.D. -Yale, in 1961 (which explains everything since) to support the conclusion that there was no difference between money and liquid assets (new money substitutes). It was based upon the following assumptions:
(1) That monetary policy has as an objective a certain level of spending for N-gDp (sound familiar?, viz., N-gDp targeting?), and that a growth in time/savings deposit classifications will not, per se, alter this objective. And that a shift from demand to time deposits will also not, per se, alter this objective;
(2) That a shift from demand to time deposits involves a decrease in the demand for money balances and that this shift will be reflected in an offsetting increase in the velocity of money.
(3) To prevent the increase in Vt from altering the desired level of spending for N-gDp, it is necessary for the FRB-NY trading desk to prevent the diminished money supply brought about by the shift from demand to time deposits from being replenished through an expansion of bank credit;
(4) To prevent the expansion of bank credit requires that the trading desk “mop up” al excess reserves created by the shift from demand to time deposit classifications.
With the monetization of time deposits (end of gate keeping restrictions), as hypothesized from 1961 until 1981 Dr. Chandler was correct.
25. February 2021 at 10:19
bb, Not sure, they are both quite slow.
Michael, There’s no reason to expect wage growth to track NGDP growth, at least if employment is growing.
Stefan, I expect NGDP targeting to happen gradually, not al lat once.
25. February 2021 at 10:38
Good piece by James Gwartney here;
https://www.aier.org/article/yes-this-time-well-have-inflation-and-heres-why/
He agrees with Scott that the paying of interest on excess reserves after 2008 kept the money supply from growing, but that we are in a new ballgame now:
—————-quote————–
…worldwide interest rates declined to historic low levels during the decade following the Great Recession. A variety of factors caused this. One that has been largely ignored was a dramatic demographic shift in high-income countries, as the number of people in the lending phase of life (roughly age 50 to 75) increased relative to those in the borrowing phase (under age 50). The resulting low and declining interest rates reduced the opportunity cost of holding money, causing the velocity or turnover rate of money to plummet. As the result of this combination of factors, the Fed’s huge increase in purchases of financial assets and money creation has, to this point, exerted only a minimal impact on inflation.
However, this favorable scenario is about to reverse course. Three major factors underlie the reversal.
First, the Fed’s current money creation dwarfs those of recent decades. Propelled by the $3.6 trillion Covid-19 spending financed by borrowing from the Fed, the narrow measure of the money supply known as M1 has expanded from $4.0 trillion to $6.8 trillion during the past 12 months, a 70 percent increase. By way of comparison, the 12-month increases in M1 during the inflationary 1970s never exceeded 10 percent. The largest previous single-year M1 increase in recent decades was a 21 percent figure in the aftermath of the Great Recession. The story is the same for the broader M2 measure of money, which has increased by 25 percent during the past year. The next largest 12-month expansions in M2 during the past 75 years were the 1975-1976 increase of 13.8 percent during the double-digit inflation of the 1970s and the 10.3 percent increase during 2011.
One has to go all the way back to World War II to find anything comparable to the money supply increases of the past 12 months. Moreover, even these gigantic figures understate the current monetary surge. The Treasury is currently holding more than a trillion dollars of committed funds in its bank account, which will be added to the money supply when they are spent in the next few months. Congress is expected to add additional fuel to the fire with the $1.9 trillion spending package currently under consideration.
Second, the inflation triggered by the huge monetary expansion will increase the expected rate of inflation and nominal interest rates. In turn, the higher nominal interest rates will cause the velocity of money to increase, unleashing additional inflationary pressures from the rapid money growth of 2008–2019. Rising inflation rates and higher nominal interest rates are two peas in the same pod. When the former occurs the latter will follow. As inflation pushes nominal interest rates upward, the recent reductions in velocity will reverse, adding to the inflationary pressure.
—————endquote————–
He predicts something like what happened after WWII ended. A return of consumer normality (pent-up demand exploding) and higher inflation. I think it is already happening. I’ve noticed that the food I usually buy at supermarkets is up about 20% in price from just a few months ago.
25. February 2021 at 22:51
Scott,
You replied:
“Michael, There’s no reason to expect wage growth to track NGDP growth, at least if employment is growing.”
Yes, that’s why I added that bit at the end about full employment:
“Am I missing something, or should workers demand wage growth equal to NGDP growth, at least at full employment?”
I’m wondering though… Can the fact that workers are often so willing to accept falling compensation/NGDP reflect underemployment, and just low labor demand, on average, or has inflation been misinterpreted?
This is a crazy thought, and likely wrong, but what if that 2-2.5% inflation rate that keeps NGDP growth above the growth in wages is a favorable trade-off in an unexpected way, such as higher prices for increasing quality of goods and services? And what if inflation over and above that rough range is excessive and harmful, if persistent?
Empirically, this reminds me that Jason Smith has shown, quite convincingly, that the quantity of money theory explains inflation rates of 10+% extremely well, but not so great for lower rates of inflation. If low levels of inflation have been misinterpreted, as I speculate, this, along with more trouble distinguishing signals from noise, could help explain why the quantity of money theory isn’t so impressive for single digit inflation rates.
Also, Jason’s model seems to forecast macro variables, such as GDP, inflation, etc. better than mainstream economic forecasts, and one way it does so is by letting the exponents involving alpha sum to more than 1, in the Solow model.
This would seem to indicate that models of productivity, such as TFP, are incomplete.
26. February 2021 at 09:12
Patrick, Inflation will rise, the question is how much. I still expect inflation to average 2% during the 2020s.
Michael, It would probably make more sense to compare wages to national income, which has risen more slowly than NGDP.
26. February 2021 at 13:45
To my way of thinking, the very term “AIT” says that the monetary authorities fundamentally do not understand the function of monetary policy. I suspect the Fed to be content with higher inflation during booms and lower inflation during busts, while claiming they’ve met the objective of average inflation.
Scott, this must be frustrating. Maybe you need to speak more loudly since you’re dealing with idiots.
26. February 2021 at 18:36
Scott,
Am I looking at the wrong national income data here, which seems empirically little different from NGDP?
https://fred.stlouisfed.org/graph/?g=Bq5m
27. February 2021 at 08:44
dtoh, You said:
“Maybe you need to speak more loudly”
All caps?
Michael, They are correlated, but NGDP tends to grow a bit faster, as depreciation is a rising share of NGDP, and not included in national income.
27. February 2021 at 14:28
The yield curve is steepening (the interest rate differential in yields between the long-end segment of the Treasury curve and the short-end segment of the Treasury curve is widening).
But some like it hot, as nominal Treasury yields remain below inflation expectations (so what’s up with the forecasts?) — all before the next stimulus. I.e., N-gDp targeting is still working.
10 yr TIPs @ -.60%
10 yr breakevens @ 2.15%
10 yr Treasuries @ 1.54%
27. February 2021 at 15:18
Scott,
I was thinking you should be less diplomatic in your criticisms of the Fed.
28. February 2021 at 10:28
dtoh, Right now the Fed’s doing OK, under the circumstances. If TIPS spread start falling sharply I may become more pointed.
28. February 2021 at 18:11
Scott,
Thanks for the tip about national income. Perhaps with enough data, it could be established that looking at the growth in such metrics versus labor compensation could be a lagging indicator of output gaps/above NAIRU unemployment. Even a lagging indicator would be better than none at all.
I’ve never seen an economist talk about the rate at which labor compensation falls relative to NGDP. After the Great Recession, that rate of decline never reached the rate that occurred after the 2001 recession, and hence never returned to the previous path.
https://fred.stlouisfed.org/graph/?g=BsRt
https://fred.stlouisfed.org/graph/?g=BsRC
This is consistent with the argument I’ve been making for around 4 years now, that something fundamentally changed after the Great Recession.
Sure, we can focus on lower NGDP growth and acknowledge some demographic factors, shifts to service sector jobs, etc., and I think a lack of demand has persisted. But why didn’t wages adjust more? Perhaps due in part to higher minimum wages, generous unemployment benefits, high effective marginal tax rates created by welfare programs? Could part of it also be some maximum adjustment rate for wages under some circumstances?
I’m playing with ways to model this.