Stagflation: What Really Happened in the 70′s
by Benjamin Studebaker
If you argue long enough about economics, you are bound to run into the stagflation argument. The stagflation argument claims that the big state and stimulus caused high inflation, high unemployment, and poor growth during the seventies. Usually this argument is not fully argued by those who believe in it–it is merely asserted, and the rest of us are expected to accept that it is simply the case that the seventies happened that way. Today I’d like to endeavour to illustrate what actually happened in the seventies, what the real causes of stagflation were, and what sort of lessons might be pulled from it.
The stagflation problem is a complex one, and we will need to refer to several Trading Economics graphs:
Inflation:

Unemployment:

Oil Prices, adjusted for inflation:

The Interest Rate:

The US Growth Rate:

By taking these four elements together, we can explain the stagflation phenomenon. As I describe what happens, follow my reasoning in the charts to confirm the validity of my analysis:
We see that the problem begins in 1973 with the 73-75 recession–that’s when growth first dives. In October of 1973, the Organisation of Petroleum Exporting Countries declared an oil embargo upon the supporters of Israel–western nations. The 73-75 recession begins in November of 1973, immediately after. During normal recessions, inflation does not rise–it shrinks, as people spend less and prices fall. So why does inflation rise from 73-75? Because this recession is not a normal recession–it is sparked by an oil shortage. The price of oil more than doubles in the space of a mere few months from 73-74. Oil is involved in the manufacturing of plastics, in gasoline, in sneakers, it’s everywhere. When the price of oil goes up, the price of most things go up. The spike in the oil price is so large that drives up the costs of consumer goods throughout the rest of the economy so fast that wages fail to keep up with it. As a result, you get both inflation and a recession at once.
So how does the government react? Upon seeing the poor growth figures, the Federal Reserve drops the interest rate in 1973 from around 11% to around 9%–a full 2% drop. Rather than buoying growth, however, this sends the inflation into overdrive. Wages cannot keep up with the subsequent spike, so purchasing power falls and growth does not return. Terrified by the double-digit inflation rate in 1974, the Federal Reserve switches gears and jacks the interest rate up to near 14%. This snuffs out a nascent economic recovery in 1974–the high rate of inflation was being factored into salaries, and consumers were regaining their spending footing. By raising rates again, the Federal Reserve begins to starve inflation back down, but does so at a cost of a very contractionary, very high interest rate. The economy slips back into the throws of the recession for another year or so, and the unemployment rate takes off, rising to around 9% by 1975.
While the economy gets out of the recession in 75′, the high unemployment hangs around. Seeking to fend off the high unemployment, the Federal Reserve drops the interest rate from 75-76, bottoming out at 5% from its double digit peak and remaining quite low through 78′. This succeeds in pulling unemployment back down to 6%, but inflation starts to trend back upwards.
Then, in 1979, the economy gets another oil price shock (this time caused by the Revolution in Iran in January of that year) in which the price of oil again more than doubles. The result is a fall in growth and inflation knocked all the way up into the teens. The Federal Reserve tries to fight the oil-driven inflation by raising interest rates high into the teens, peaking out at 20% in 1980. Inflation responds, sagging back to around 10%. But these high interest rates again drive the economy back into recession through the first half of that year and meanwhile, unemployment jumps up to just under 8%. The Federal Reserve responds by backing off substantially on rates in mid to late 1980, and the result is a brief return to growth in 1981. But when inflation again begins rising from 10%, the Federal Reserve returns to high rates in 1981, keeping rates above 15% into 1982.
This, combined with oil prices sagging back down from their peak, breaks the back of inflation, which, by 1982, has slid down to acceptable levels. Meanwhile, however, the economy is back in recession and, by 1983, the unemployment rate has peaked at nearly 11%. To fight this, the Federal Reserve knocks the interest rate back below 10%, and meanwhile, alongside all of this, Ronald Reagan spends lots of money and expands the state in 82/83 (a nice graph of this data conveniently puts Obama’s spending numbers alongside Reagan’s–this is extraneous to our point, but the reader may nonetheless find it interesting):

Why does inflation not respond by returning? Because oil prices are falling throughout this period, and, by the end of most of our graphs (1985), it has collapsed utterly.
So what we see here really is not any kind of complicated challenge to the notion that expansionary Keynesian policy is expansionary and contractionary Keynesian policy is contractionary. What we instead see is a crisis in the price of a very necessary good, namely oil, which creates an unacceptably high basic inflation rate and makes doing all kinds of business prohibitively more expensive.
All the normal Keynesian rules applied during the seventies–higher interest rates meant less inflation and more unemployment, lower rates meant less unemployment and more inflation. The only difference was that the baseline for inflation was much higher–the choice was between insane inflation and high inflation, not between high inflation and some kind of normal or acceptable level. Ultimately in both the mid-seventies case and the early eighties case, inflation only comes down permanently from its high baseline when the economy has adjusted to the higher oil price as a new normal and/or when the oil price itself comes down.
What would have prevented us from enduring the misery of the seventies? Less economic dependence on oil, both in terms of energy and throughout the wider economy. Jimmy Carter made a pitch for it, though the country declined to take him up on it at the time.
The issue was never one of needing to switch from stimulus spending to tax cuts for the rich, or of needing to care exclusively about inflation and ignore unemployment. People who interpreted the seventies in that fashion misunderstood fatally what was going on. That misunderstanding fuels and feeds their misunderstanding of the present problems.
There is not going to be hyperinflation or stagflation or any of the various seventies style horrors the economic right likes to conjure in its criticisms of stimulus and Keynesianism, because there is no giant spike in any ubiquitous, necessary thing like oil. Ronald Reagan dropped the interest rate and increased state spending in 82/83, and that, in tandem with falling oil prices, is what brought about “morning in America”. Those are Keynesian tactics; there’s no paradigm shift here. The conventional wisdom about this period in our history is just wrong–to the extent that Reagan did good economically, he did good not because of some new conservative economic agenda, but because he followed Keynesian orthodoxy and had the good fortune to see oil prices fall during his presidency.
The only lesson that can be taken from the seventies is that it is a very foolish thing for a state to become dependent on a particular resource or product that it cannot make for itself and must import, because that puts the economic welfare of that state at the mercy of the exporting nations. No matter how powerful a country is, if it is dependent in this way, otherwise small, powerless nations can bring it to its knees.
This analysis misses the point. Yes, the world economy suffered supply-side inflation shocks. But the point is that – under Keynesian regimes – these shocks triggered persistent inflation due to inflationary expectations incorporating policy accommodation. This led to a bad equilibrium of high unemployment/high inflation – i.e. stagflation.
It was only when Volcker committed the Federal Reserve to a hawkish stance that inflation came under control as markets reduced their inflation expectations to incorporate the expectation of future policy. In contrast, Keynesian policy neglected the supply side in favour of the demand side, so it was more vulnerable to inflation shocks and weak with dealing with them.
The Federal Reserve was active in its efforts to reduce inflation long before Volcker, as we can see in the 73-75 period. It only moved from those high rates because the rate of inflation was on its way down (to under 6% in 77′) and because in the meantime, the unemployment rate was climbing and growth was poor. It was the Iranian Revolution’s second shock to oil prices, not a Keynesian indifference to inflation, that sent inflation back up so very high again in 79′.
Again, missing the point: the point is that inflation became persistent as markets believed the oil shocks would become accommodated. This accommodation can be seen in the 1970s through the extent to which interest rates were reduced after reaching their early peak. Given that this accommodation was expected (and realised) workers bargained for higher wage increases and firms increased prices – turning a temporary shock into persistent inflation. Contrast this the Fed policy in the early 80s where interest rates were kept high for a prolonged period. As workers believed this hawkishness was credible, they bargained for lower wage increases accordingly.
If your argument were correct, we would see a spike in the labour share of GDP during the seventies over and above that of the fifties and sixties. No such spike exists in the BLS data:
http://www.nakedcapitalism.com/wp-content/uploads/2012/12/120912krugman1-blog480.jpg
You’d see a spike in the labor share of income, which can be seen here:
http://i50.tinypic.com/346u7vr.jpg
Whose data is this? I used the impartial Bureau of Labour Statistics; this provides no source.
National Income and Product Accounts (NIPA). Labour share obtained by dividing total compensation in the nonfinancial corporate sector by nominal nonfinancial corporate output.
Do we have a reason to take NIPA’s word over that of the BLS? Is their methodology demonstrably better?
If NIPA and the BLS are equally credible, then we average their data sets and come up with a negligible small rise in labour share during the seventies, with periods of the seventies still resembling the fifties and sixties.
Most importantly, even if NIPA is more credible, it is just as reasonable, if not moreso, to argue that changes in the labour share reflect the other changes I discuss as it is to say that they caused them. The NIPA figure for labour share rises with high inflation and low unemployment and falls with low inflation and high unemployment during the seventies. That is intuitively sensible–when oil prices deliver a shock to supply, driving up inflation and cutting into the profits derived from capital, labour is left to gain. When the Federal Reserve reacts to reduce inflation, unemployment is generated, and the labour share falls back off, even under the NIPA data. The two huge hikes in labour share under the NIPA data match the two oil price spikes (and subsequent inflation spikes) in 73 and 79 and the two huge plunges in labour share under the NIPA data match the two periods of high unemployment, in the mid to late seventies and early eighties. It is perfectly possible–indeed I would argue, even probable–that the labour share was responding to the macroeconomic forces of the period, not the reverse.
Interesting post! What was happening with wages growth and the labour/capital share of the pie over this period? From what I’ve gathered from my limited understanding of the period its not just Keynesian statism that was blamed by Conservatives but also the intransigence of organized labor.
A fantastic question! I found some Federal Reserve statistics that include the period:
http://www.clevelandfed.org/research/Commentary/2012/2012-13-1w.gif
We see the labour share declining during the 73-75 recession, recovering in the late seventies, and then diving back down during the early eighties recession, never again to achieve as high a percentage as that posted during the turn of the decade. The labour share is much lower now than during the seventies, but it should be noted that the share in the seventies was no greater than it was during 40′s and 50′s, the apex of the post-war boom, if you trust the BLS over the NIPA, as I am inclined to do.
Thanks! Looks as though the “wage inflation” angle can only be defended in the leadup to the 69/70 recession, and even then only on the basis of the NIPA data. The only significant pattern is that labor’s share of income stagnated after the 70s and then precipitously dropped during the 00′s.
Are you in a position to manual me on your web marketer or even the man whom manages your blog post, I’d like to determine it might be easy to be a guest poster.
I run the entire show here all on my own. One man’s work, with the aid of WordPress.