Misremembering history

The Week has an article by Jeff Spross, which tries to rewrite the history of the early 1980s:

The story of how Volcker fulfilled his mission is complicated. But it boiled down to a massive hike in interest rates: The Fed’s primary target for those rates reached an astronomical 19 percent in 1981.

Actually, during 1979-82 the Fed switched from targeting interest rates to targeting the money supply.  (Although there is debate about whether they actually were targeting the money supply.)

The basic problem, as economist Dean Baker explained to The Week, is there’s no way to tame inflation that doesn’t involve inflicting damage on the economy. But using interest rate hikes to spark recessions is a methodology that loads the bulk of the pain onto everyday workers, and people who are marginalized in our society. The national unemployment rate (the blue line below) briefly reached 10.8 percent — higher than it got even in the Great Recession — and it didn’t get back to 5 percent until 1989. Which was bad enough. But the unemployment rate for lower class workers is always much higher than for upper class ones. Ditto racial minorities: The unemployment rate for African-Americans (the red line below) topped 20 percent by 1983.

And that’s not all:

The Volcker recession also roughly coincides with a remarkable inflection point in the American economy. Before the mid-1970s, labor markets were often tight and full employment was common. After the Volcker recession, full employment — when there are more jobs available than workers, so employers have to bargain up wages and work conditions — basically disappeared. Union membership had already fallen 5 percentage points from roughly 1960 to 1980. But after the Volcker recession, its decline accelerated, falling another 10 percentage points from 1980 to roughly 1995.

Most strikingly, the Volcker recession falls almost right atop the moment when inequality took off:

So what’s the alternative?

All of that raises the obvious question: Could we have done things differently?

The 1970s were actually a relatively straightforward version of standard Keynesian macroeconomic theory, in which an overheating economy drives up the inflation rate. The real economic growth rate was actually quite strong that decade, bouncing around near 5 percent.

This alternative doesn’t get off to a promising start.  Real growth was closer to 3%, not 5%.  And even that was mostly due to very fast labor force growth (boomers like me, plus women entering in large numbers).  In fact, the 1970s was the decade when the Keynesian model basically collapsed, and had to be replaced by New Keynesianism, which, as Brad DeLong later pointed out, was to a substantial extent monetarism.

Keynesianism says that hiking taxes and balancing the budget, or even driving it into surplus, will put a drag on economic growth and slow down inflation. So we could’ve raised taxes and returned to the mid-century model of lots of brackets and really high rates at the top. That would’ve inflicted most of the damage on richer Americans who can afford to absorb it, rather than the workers and the poor who can’t.

Where to start?  Even if this were true, the old Keynesian (Phillips curve) model says the slowdown in inflation would have produced mass unemployment, regardless of whether it was caused by fiscal austerity or tight money.  And of course the model is not true.  LBJ tried to control inflation with tax increases in 1968, and failed.  That was one of the key policy experiments (along with Volcker’s later success in controlling inflation, during a period of fiscal stimulus under Reagan) that led economists to abandon old Keynesianism. The theory simply does not work.  Even worse, the left-wing solution of high taxes was tried in Sweden during the 1970s and 1980s, and had to be abandoned in the early 1990s, as Sweden was rapidly losing ground relative to other developed countries.

On top of that, inflation was exacerbated by several historical flukes. The most obvious was the rise in oil prices over the 1970s, driven by OPEC’s oil embargo against the U.S., and then by the shutdown in Iranian oil exports brought on by the Iranian revolution.

This is a common myth, at least for the decade as a whole (perhaps true for 1974 and 1979).  During the period from 1972-81, NGDP growth averaged 11%, divided into 3% RGDP growth and 8% inflation.  Oil shocks may impact inflation, but they don’t impact NGDP.  Even if we had had no oil shocks during the 1970s, an 11% NGDP growth rate was enough to generate 8% inflation.  Even with 4% RGDP growth, inflation would have averaged 7%.

All these problems would’ve likely worked themselves out regardless of what Volcker did. The error in the CPI was corrected in 1982. The run-up in oil prices from the Middle East drove a massive surge in oil production in other parts of the world, and a big uptick in energy efficiency, all of which would’ve naturally pushed inflation back down. And as mentioned, unions were losing their clout, with membership already on a downslide.

“We could’ve expected inflation to fall in any case,” Baker continued. “Probably not as quickly and not as much.”

“But suppose it took us 6 years to get down to 4 percent inflation? What would’ve been the problem with that?”

The real problem was different; the short 1980 recession was a complete waste, accomplishing nothing.  Volcker got cold feet in the summer of 1980 and adopted an extremely expansionary monetary policy (the Fed was worried about Carter’s chances in the fall election).  For those who like “concrete steps”, the Fed drove real interest rates sharply negative in mid-1980.  This pushed NGDP growth up to an astounding 19.2% in 1980:4 and 1981:1.  Thus the Fed had to tighten again in mid-1981.  The recession from mid-1981 to the end of 1982 should have started at the beginning of 1980, in which case it would have been over by mid-1981.  And it would have been milder than the actual 1981-82 recession, because inflation expectations were even more deeply entrenched after the 19.2% spurt in NGDP after the brief 1980 recession.

Plus, Ronald Reagan would have looked far more heroic if the recession had ended in mid-1981.


Tags:

 
 
 

22 Responses to “Misremembering history”

  1. Gravatar of Dimitri Klimenko Dimitri Klimenko
    21. April 2016 at 10:49

    Have you considered the idea that a major cause of the 1970s/80s inflation was the onset of financial innovation in the form of credit cards, and later debit cards?

    If the value or ‘price’ of cash derives (per your “hot potato” argument) primarily from its ability to be used in transactions, shouldn’t it follow that when competing transactions technology is developed the price of cash must fall?

    The monetarist response of constricting the monetary base seems like the correct one, but given an unprecedented drop in the ‘price of money’ I can see how central banks might have taken a while to get to grips with the situation.

  2. Gravatar of Don Geddis Don Geddis
    21. April 2016 at 11:38

    @Dimitri Klimenko: “If the value … of cash derives … primarily from its ability to be used in transactions

    No, the primary effect of money on the macro economy, is through its use as the unit of account, not the use as the medium of exchange. Changes to transaction technology don’t have macro significance, if they leave the unit of account unchanged.

  3. Gravatar of ssumner ssumner
    21. April 2016 at 11:44

    Dimitri, The facts don’t fit that theory. You can explain the inflation by just looking at the growth in the monetary base. Of course as inflation rose sharply, the opportunity cost of holding base money increased, and hence velocity rose somewhat as well. But it all comes down to printing money. If they had printed money at a slower rate, there would have been no Great Inflation.

  4. Gravatar of collin collin
    21. April 2016 at 13:11

    Reading this post, the inflation of the 1970s was primarily because we had sharp increase in labor supply due to the Baby Boomer reaching adulthood and women in the workforce without lowering wages to fix markets. I believe equal rights and wages for minorities played a role as well. Oil shocks were simply added ‘gasoline’ to the fire.

    With Reagan and the 1980s, real wages finally fell and outside the dotcom boom years, have continued so. (The median household income rose during the 1980s due to more 2 income families.) Until 2010, when the labor supply starting falling due to increased focus on human capital for young people and retirement started for Boomers. I still the impact of changing demographics has more influence on the slowdown of the global economy.

  5. Gravatar of Art Deco Art Deco
    21. April 2016 at 13:33

    Have you considered the idea that a major cause of the 1970s/80s inflation was the onset of financial innovation in the form of credit cards, and later debit cards?

    Consumer credit was not an innovation of the 1970s. People had store accounts. That aside, charge cards appeared in 1951 and credit cards in 1968. The Fed successfully re-stabilized prices in 1951/52 and they remained roughly stable until Wm. Martin caved in to LBJ in mid 1966.

  6. Gravatar of Art Deco Art Deco
    21. April 2016 at 13:46

    Reading this post, the inflation of the 1970s was primarily because we had sharp increase in labor supply due to the Baby Boomer reaching adulthood and women in the workforce without lowering wages to fix markets. I believe equal rights and wages for minorities played a role as well. Oil shocks were simply added ‘gasoline’ to the fire.

    Birth cohorts rose rapidly in size until about 1930, at which time they bounced around a set point just north of 2 million for nine years and then began moving upward in a jagged fashion. The increase in birth cohort sizes was nearly complete by 1952 (though not absolutely complete until 1957). The increased flow into the labor market would have been ongoing from about 1957 to about 1970. The inflation took off in 1966, so your conception is anachronistic at the least. It should be noted that the ratio of employed persons to the total population of persons over the age of 16 stood hardly changed between 1957 and 1970 (standing at 0.57 at both times).

  7. Gravatar of Art Deco Art Deco
    21. April 2016 at 13:48

    real wages finally fell and outside the dotcom boom years, have continued so.

    They didn’t and they haven’t.

  8. Gravatar of TravisV TravisV
    21. April 2016 at 14:13

    “Its biggest headache may be a nasty spat with Berlin after Finance Minister Wolfgang Schauble said its policies were causing “extraordinary” problems for Germany and were in part to blame for the rise of the right-wing anti-immigration Alternative for Germany (AfD).”

    http://www.businessinsider.com/r-draghi-keeps-rates-at-rock-bottom-in-face-of-german-backlash-2016-4

    http://www.businessinsider.com/european-central-bank-press-conference-april-21-2016-4

  9. Gravatar of Benjamin Cole Benjamin Cole
    21. April 2016 at 15:16

    Scott–Reagan was also the greatest protectionist since Hoover and Smoot-Hawley. Milton Friedman said Reagan made Smoot-Hawley look tame!

    Niskanen points out the range of goods subject to tarriffs grew from mid-teens to mid-upper twenties under Reagan ( though we must begrudgingly admit that Reagan may have saved our American Icon Harley-Davidson).

    Indeed, what Reagan actually did makes what Trump proposes look rather feeble.

    Yet the US economy expanded under Reagan. So are Trump’s policies misguided but actually not that dangerous?

  10. Gravatar of Dimitri Klimenko Dimitri Klimenko
    21. April 2016 at 18:01

    @Don
    I don’t really buy the “unit of account” argument. If a bunch of banks started to keep loans and deposits in units of CPI baskets rather than yen, I don’t think this would have all that much effect on the price level as long as people continued to make transactions in, and pay taxes in, USD.

    @Scott
    I’m quite skeptical of the idea that the monetary base alone explains everything. The monetary base doubled from 1970 to 1980, doubled again from 1980 to 1990, and again from 1990 to 2000. It also tripled from 2008 through to 2011. Meanwhile, it’s only the 1970s that saw massive inflation.

    It’s not good enough to resort to “expectations”, because unless it can properly anchor those expectations your theory has lost all explanatory power, and could be used to explain *anything*.

    The other issue I have with the “it’s the monetary base!” argument is that the inflation of the 1970s and 1980s was a worldwide phenomenon. The Federal Reserve can clearly be blamed for making it better or worse, but the underlying explanation has to be *global*.

    @Art Deco,
    Yeah, that’s a good point. However, innovation is not the same thing as invention. After all, the economic value of the technology only comes about when you get widespread adoption.

  11. Gravatar of Art Deco Art Deco
    21. April 2016 at 18:53

    it’s only the 1970s that saw massive inflation.

    Averaged about 7% a year. Annoying, not massive.

  12. Gravatar of ssumner ssumner
    21. April 2016 at 19:08

    Collin, You said:

    “Reading this post, the inflation of the 1970s was primarily because we had sharp increase in labor supply due to the Baby Boomer reaching adulthood and women in the workforce without lowering wages to fix markets. I believe equal rights and wages for minorities played a role as well. Oil shocks were simply added ‘gasoline’ to the fire.”

    Just the opposite. Without the large increase in the labor force, inflation would have been even higher. It was caused by the Fed printing too much money.

    Travis, The Germans are not being helpful.

    Ben, You said:

    “So are Trump’s policies misguided but actually not that dangerous?”

    No, his policies are extremely dangerous. Fortunately they won’t be enacted.

    Dimitri, You said:

    “I’m quite skeptical of the idea that the monetary base alone explains everything.”

    Saying the monetary base explains everything, is not the same as saying the inflation rate equals the monetary base increase. When inflation accelerates, base velocity rises, and vice versa. That explains most of what you cite there. Of course the period since 2008 is different, due to the zero bound.

    As fas as the global phenomenon, the various inflation rates in different countries are highly correlated with the monetary base growth rates. Check this out:

    https://www.themoneyillusion.com/?p=30081

  13. Gravatar of Dimitri Klimenko Dimitri Klimenko
    21. April 2016 at 21:00

    Scott, You said:
    “Saying the monetary base explains everything, is not the same as saying the inflation rate equals the monetary base increase. When inflation accelerates, base velocity rises, and vice versa. That explains most of what you cite there.”
    As you know, “base velocity” is just a definition, it doesn’t have real explanatory power. Yes, when inflation is high, issues such as price stickiness cause people to shift their purchases forwards in time, thus amplifying the effects of that inflation. That is, however, a short-run effect.

    I don’t think the long-run QTM effect is a sufficient explanation for the two decades from 1970 to 1990. Over that period of time, the monetary base increased by a factor of around 3.9, while nominal GDP increased by a factor of around 5.5: the discrepancy is huge. It’s also a poor explanation for what triggered the inflation in the first place; if it was due solely to bad monetary policy, why would so many other countries experience the same phenomenon?

    Scott, You said:
    “Of course the period since 2008 is different, due to the zero bound.”
    The “zero lower bound” is a special property of New Keynesian models, but what does that have to do with the predictions of QTM? I don’t think that excuse holds up.

    Scott, You said:
    “As fas as the global phenomenon, the various inflation rates in different countries are highly correlated with the monetary base growth rates. Check this out:

    https://www.themoneyillusion.com/?p=30081
    I agree completely that there is generally a strong correlation between growth in the monetary base and growth in nominal GDP.

    Where I disagree with you is in the assertion that the long-run causal relationship from monetary base to NGDP is one-for-one, i.e. that if you gradually double the monetary base over some period of time you expect to see a doubling in NGDP.

    Historically, when you pretty much had to use cash money for every transaction, this was probably true. However, in a modern world which is increasingly cashless, I simply don’t think naive QTM holds up any longer.

  14. Gravatar of Dimitri Klimenko Dimitri Klimenko
    21. April 2016 at 21:20

    Also, the economic argument is a pretty simple one. According to QTM, all of the value of money must come from the fact that it enables you to make trades (including state-contingent and intertemporal trades) that you could not make without it.

    However, this also means that this source of economic value is an untapped resource, and the principles of market economics dictate that the finance industry will compete to seize as much value as they can by making transactions more efficient.

    We live in a world where the “inside money” of banks and the finance industry has driven the marginal transactions-value offered by base money to nearly zero. Consider, for example, that many countries currently have banks whose reserves are sitting there earning *negative* interest, because the costs of dealing with cash already outweigh its benefits.

    It’s not the zero lower bound that’s the issue; it’s simply that we live in a world where QTM doesn’t really apply any more. Basically, you can think of QTM and chartalist (e.g. FTPL) theories of the price level as representing two extremes:
    1) QTM applies to a world where government-issued “base money” has a monopoly on transactions services in the economy.
    2) Chartalist theories apply to a world where financial competition acts to drive the marginal costs of transactions services down to nearly zero.

  15. Gravatar of Dimitri Klimenko Dimitri Klimenko
    21. April 2016 at 22:05

    Perhaps an even simpler way to put it: “moneyness” is a *service*; it allows you to make transactions you would not be able to make (or would have much more difficulty making) without it.

    Under QTM, the government has a monopoly on that service and thus can unilaterally set the price level.

    In the real world, the private sector competes with the government for provision of the “moneyness” service, which acts to drive down the price of moneyness regardless of what the government does.

    With enough competition, the price of government money necessarily falls towards a “base level” which reflects services where the government has either a de facto or a de jure monopoly.

    In particular, base money has a de jure monopoly over payment of taxes, and a de facto monopoly (shared with government bonds) for insurance against major economic crises.

  16. Gravatar of Ray Lopez Ray Lopez
    21. April 2016 at 23:16

    Dimitri Klimenko wins this thread. What he says is uncontroversial, mainly that credit cards destroyed tracking M1, yet Sumner insists on trying to rebut him. Sad.

    And this: “Actually, during 1979-82 the Fed switched from targeting interest rates to targeting the money supply. (Although there is debate about whether they actually were targeting the money supply.)” (Sumner) – actually, if you look at what the US Fed did and how the economy responded, you’ll see when the Fed RAISED interest rates, the inflation rate ROSE, while when the Fed LOWERED interest rates, the inflation rate FELL, the OPPOSITE of monetary theory. Money is endogenous, money is neutral. Unless you are welling to constructed complicated Ptolemy-type epicycle-equivlaents for the economy, with complicated variable lag/lead effects that change from time to time. Much more logical to assume money is largely neutral, which the evidence upholds (Bernanke’s 2002 FAVAR paper is good start).

  17. Gravatar of ssumner ssumner
    22. April 2016 at 05:40

    Dimitri, You said:

    “As you know, “base velocity” is just a definition, it doesn’t have real explanatory power. Yes, when inflation is high, issues such as price stickiness cause people to shift their purchases forwards in time, thus amplifying the effects of that inflation. That is, however, a short-run effect.”

    Of course it has explanatory power–there are 100s of money demand studies that model velocity. And I don’t know what you mean by shifting purchases forward in time. Are you describing a change in consumption/saving preferences? Then say so. But that’s a different issue from velocity, which depends on nominal interest rates, not real rates. Real interest rates were low during the 1970s, so any move away from saving and toward consumption was not powerful to significantly impact velocity. The speed up in velocity was caused by higher inflation, which was caused by a dramatic increase in the rate of growth in the monetary base after 1964.

    You said:

    “It’s also a poor explanation for what triggered the inflation in the first place; if it was due solely to bad monetary policy, why would so many other countries experience the same phenomenon?”

    You need to check out my link.

    And the difference between base growth and NGDP growth between 1970 and 1990 is consistent with theories of money demand. Income tax rates fell between 1970 and 1990, so there was less currency hoarding. Money demand is not a constant, it’s a function.

    You said:

    “Where I disagree with you is in the assertion that the long-run causal relationship from monetary base to NGDP is one-for-one, i.e. that if you gradually double the monetary base over some period of time you expect to see a doubling in NGDP.”

    I’ve never made that claim, and I don’t think anyone else has. The claim is that doubling the base causes NGDP to be twice as high as if the base was not doubled. But there are many other factors involved, including the rate of change in M and P. You are oversimplifying all this to a rather extreme extent. I’m not saying that 5% base growth in the 1972-81 period would have produced 5% NGDP growth, instead of 11%, I’m saying it would have produced something much lower, say 4% or 6%, instead of 11%.

    You said:

    “However, in a modern world which is increasingly cashless,”

    This is simply false, Cash as a share of GDP in the US is higher than in the 1920s. Printing money does not cause inflation because money is used in transactions, it causes inflation because money is the medium of account, and people care about the real quantity of money they hold, not the nominal. Under the gold standard, a major discovery of gold is inflationary even if gold is not used in transactions.

    You said:

    “Also, the economic argument is a pretty simple one. According to QTM, all of the value of money must come from the fact that it enables you to make trades (including state-contingent and intertemporal trades) that you could not make without it.”

    False, for the reasons explained above. But even if true, it would not have the implications you assume, unless there are perfect substitutes for cash. And there are not.

    You said:

    “We live in a world where the “inside money” of banks and the finance industry has driven the marginal transactions-value offered by base money to nearly zero. Consider, for example, that many countries currently have banks whose reserves are sitting there earning *negative* interest, because the costs of dealing with cash already outweigh its benefits.”

    You are not thinking at the margin. It’s costly to store large amounts of cash, that’s true, but small amounts are still very useful in transactions.

    Ray, You said:

    “if you look at what the US Fed did and how the economy responded, you’ll see when the Fed RAISED interest rates, the inflation rate ROSE, while when the Fed LOWERED interest rates, the inflation rate FELL, the OPPOSITE of monetary theory.”

    You’ve never heard of the Fisher effect? It’s good to see you haven’t lost your ability to say things that make me laugh. Thank’s for bringing a smile to my face.

  18. Gravatar of Craig Craig
    22. April 2016 at 08:22

    Scott, is the a good monetary history reference 1960- that you agree with? If not, is it time for your next book?

  19. Gravatar of Goose Goose
    22. April 2016 at 19:34

    Wouldn’t rising prices disproportionately “impact” the poor vs the rich as well? I don’t think the elite in venezuela or zimbabwe feel the pinch quite as much as the farmers or plumbers.

  20. Gravatar of ssumner ssumner
    23. April 2016 at 06:03

    Craig, I am working on a book, but I recommend Bob Hetzel’s book on the Great Recession.

    Goose, You need to first consider what causes prices to rise. If it’s monetary policy, then incomes will also rise.

  21. Gravatar of Harry Harry
    25. April 2016 at 16:49

    > the Fed was worried about Carter’s chances in the fall election

    This is the first I’ve heard of this. Could you point to any sources for it?

  22. Gravatar of ssumner ssumner
    26. April 2016 at 05:10

    Harry, My mistake, I should have said “in my view”. Obviously the Fed did not publicly say that. But what other excuse is there for slashing interest rates by 100s of basis points just months before the election, when inflation is running at 13%. Carter appointed Volcker.

Leave a Reply