Winner of the New Statesman SPERI Prize in Political Economy 2016

Tuesday 14 October 2014

The mythical Phillips curve?

Suppose you had just an hour to teach the basics of macroeconomics, what relationship would you be sure to include? My answer would be the Phillips curve. With the Phillips curve you can go a long way to understanding what monetary policy is all about.

My faith in the Phillips curve comes from simple but highly plausible ideas. In a boom, demand is strong relative to the economy’s capacity to produce, so prices and wages tend to rise faster than in an economic downturn. However workers do not normally suffer from money illusion: in a boom they want higher real wages to go with increasing labour supply. Equally firms are interested in profit margins, so if costs rise, so will prices. As firms do not change prices every day, they will think about future as well as current costs. That means that inflation depends on expected inflation as well as some indicator of excess demand, like unemployment.

Microfoundations confirm this logic, but add a crucial point that is not immediately obvious. Inflation today will depend on expectations about inflation in the future, not expectations about current inflation. That is the major contribution of New Keynesian theory to macroeconomics.

This combination of simple and formal theory would be of little interest if it was inconsistent with the data. A few do periodically claim just this: that it is very hard to find a Phillips curve in the data. (For example here is Stephen Williamson talking about Europe - but see also this from László Andor claiming just the opposite - and this from Chris Dillow on the UK.) If this was true, it would mean that monetary policymakers the world over were using the wrong framework in taking their decisions.

So is it true? The problem is that we do not have good data series going back very far on inflation expectations. Results from estimating econometric equations can therefore vary a lot depending how this crucial variable is treated. What I want to do here is just look at the raw data on inflation and unemployment for the US, and see whether it is really true that it is hard to find a Phillips curve.

The first chart plots consumer price inflation (y axis) against unemployment (x axis), where a line joins one year to the next. We start down the bottom right in 1961, when inflation was about 1% and unemployment 6.7%. Over the next few years we get the kind of pattern Phillips originally observed: unemployment falls and inflation rises.

The problem is that with inflation rising to 5.5% in 1969, it made sense for agents to raise their expectations about inflation. (In fact they almost surely started doing this before 1969, which may give the line from 1961 to 1969 its curvature. For given expectations, the line might be quite flat, a point I will come back to later.) So when unemployment started rising again, inflation didn’t go back to 1%, because expected inflation had risen. The pattern we get are called Phillips curve loops: falling unemployment over time is clearly associated with rising inflation, but this short run pattern is overlaid on a trend rise in inflation because inflation expectations are rising. Of course the other thing going on here is that we had two oil price hikes in 1974 and 1979. The chart finishes in 1980.

Most economists agree that things changed in 1980, as Volker used monetary policy aggressively to get inflation down. The next chart plots inflation and unemployment from 1980 to 2000.

Inflation came down from 13.5% in 1980 to 3.2% in 1983 partly because unemployment was high, but also because inflation expectations fell rapidly. (We do have survey evidence showing this happening.) The remaining period is dominated by a large fall in unemployment. So why didn’t this fall in unemployment push inflation back up? In terms of the chart, why isn’t the 2000 point much higher? Again expectations are confusing things. One survey has inflation expectations at around 5% in 1983, falling towards 3% at the end of 1999. So inflation was being held back for that reason. A Phillips curve, and its loops, is still there, but pretty flat.

The final chart goes from 2000 to 2013. Note that the inflation axis has changed - it now peaks at 4.5% rather than 16%. The interesting point, which Paul Krugman and others have noted, is that this looks much more like Phillips’s original observation: a simple negative relationship between inflation and unemployment. This could happen if expectations had become much more anchored as a result of credible inflation targeting, and survey data on expectations do suggest this has happened to some extent. There are also important changes in commodity prices happening here too.

While the change in inflation scale allows us to see this more clearly, it hides an important point. Once again the Phillips curve is pretty flat. We go from 4% to 10% unemployment, but inflation changes by at most 4%. However from the previous discussion we can see that this is not necessarily a new phenomenon, once we allow for changing inflation expectations.

Is it this data which makes me believe in the Phillips curve? To be honest, no. Instead it is the basic theory that I discussed at the beginning of this post. It may also be because I’m old enough to remember the 1970s when there were still economists around who denied that lower unemployment would lead to higher inflation, or who thought that the influence of expectations on inflation was weak, or who thought any relationship could be negated by direct controls on wages and prices, with disastrous results. But given how ‘noisy’ macro data normally is, I find the data I have shown here pretty consistent with my beliefs.



  1. "Microfoundations confirm this logic, but add a crucial point that is not immediately obvious. Inflation today will depend on expectations about inflation in the future, not expectations about current inflation. That is the major contribution of New Keynesian theory to macroeconomics."

    British nurses are on strike because their wages have not kept up with rises in other professions (including MPs.) Apparently New Keynesian economics has "shown" (whatever that means) us that inflation today depends on expectations of inflation in the future.

    I do not think we can generalise that way. There is too much uncertainty for most agents and even their representative bodies to behave that way and it leads all sorts of aggregation problems. And surely it depends on the type of labour market and national wage setting process we are talking about? In this deflationary environment now called 'secular stagnation" there are lot of other things driving the rate of increases in various goods prices besides the unemployment rate.

  2. Unemployment isn't as simple as it used to be. You have to consider "the underemployment rate". The dreaded NAIRU (non-accelerating inflation rate of unemployment) is another neo-liberal myth. The long term unemployed are disconnected from the wage / price setting mechanism. So the true unemployment for the Phillips Curve is short term unemployment to which must be added part time / "zero hours" under employment.

    This creates a considerable "buffer stock" of working hours available for employers to call on. This suppresses even the full time skilled workers who think they have a lock on wage / overtime rates. Hence, little wage push inflation for a while yet. Low skill, low pay employment with in-house training is making a come-back to add a third element to the buffer stock of "Worker Bees".

    All the best to you and yours. Acorn.

    1. What you're saying doesn't really sound like an outright rejection of the NAIRU though. Surely the concept of the NAIRU can still hold it's just that it's hard to pin down and we shouldn't judge where we're at simply by looking at headline unemployment.

    2. Andreas, I think I may be coming to the conclusion that if NAIRU ever existed and that the long run Phillips curve is vertical; it is not socially acceptable. I have come around to the MMT position that a buffer stock of employed is better than one of (neo-liberal style) unemployed. Have a look at the following and see what you think. Thanks Acorn.

  3. There is the Osborne Curve, of course, where 1.2% CPI is a triumph of his marvellous stewardship.

    1. The Osborne Constraint and NAIRU are essentially the same concept.

  4. Here's a 3D representation of US data:

    It must be completely dependent on expectations, because there doesn't seem to be much structure in the data.

  5. George Evans Old Keynesian "bottleneck" model of Phillips curve (EJ 1985) is far more insightful than any of the New Keynesian models.

  6. What I find amusing about people who deny the Phillips curve or NAIRU or any relationship between inflation and unemployment (e.g. Lars Syll and Bill Mitchell) is that they don’t jump to the obvious conclusion that unemployment can be brought down by leaps and bounds in no time at all simply by engineering a huge increase in demand. Reason is, of course, that they know perfectly well that when demand rises far enough and unemployment comes down far enough, inflation takes off.

  7. Sorry, but the article confuses all the time levels with differences. More inflation means an acceleration, not a rise in prices. And for sure, microeconomic foundations do NOT confirm the logic behind a Phillips Curve. The PC is theoretically at odds with neoclassical theory of the labor market.
    Best Regards,

  8. posted as anonymous but moniker at footer.

    Pablo is correct.
    This analysis is such a mish mash, its like you have thrown in some Krugman, Mankiw and Friedman, stirred them together and come out with Chicken Fried Krugman, with inflation fries (sometimes I like my brains swift inventions!) ...

    anyone who is new to this game and wants a better analysis just search bill mitchell (already refd above by prev posters) phillips curve.

    Ralph I see you have stopped posting over there ... did you finally tire of being shown to be incorrect and so have retreated to safer ground? I actually liked you showing up as it allowed the rest of us to pause and think (briefly) of ways to prove your illogical positions wrong. I was sorry to see you got bullied a little toward the end, i hope this didnt put you off.

    Regards. Hugh of the north.

    1. No: I haven't stopped commenting on Bill Mitchell's site. I left a comment about a week ago.

      Re my alleged "incorrectness", what incorrectness? I certainly don't go along with Bill Mitchell's bizarre idea that NAIRU is some sort of wicked plot by the elite to deliberately increase unemployment. In fact I noticed that Bill Mitchell himself used NAIRU in very much the conventional sense (e.g. as per the definition in the Oxford Dictonary of Economics) in a recent paper.

      Thus Bill Mitchell seems to be contradicting himself. Which is nice because it means there is no need for me to contradict him. I'll leave the link to the latter article here later if I can find it.

  9. And thus you prove my point - just because you SAY something doesnt make it TRUE! You cant just provide a link (you havent even done that much yet) to him using a NAIRU term and expect that to be sufficient to prove you he has used it in a conventional sense, but following this you then have to PROVE your inflammatory and ignorant statement that he believes it is a wicked plot of the elite. He has never said any such thing and if your comprehension skills were higher you would know that.

    I simply advise other readers to go the serious MMT blogs (and go to bills if you want to get a handle on who provides serious follow up comments, I believe most will conclude the obvious about Ralphs abilities)

    Hugh of the north

  10. "Inflation came down from 13.5% in 1980 to 3.2% in 1983 partly because unemployment was high, but also because inflation expectations fell rapidly. … One survey has inflation expectations at around 5% in 1983"

    This seems like putting the cart before the horse. Expectations of inflation can be drastically lowered if people see inflation drastically falling.

    Volcker manufactured a massive recession in 1982. Recessions cause a steep fall in inflation. The US 1953 recession caused deflation in 1954. The 1958 recession caused inflation to drop from 3.5% to 0.5% by mid-1959. The 1970 recession caused inflation to fall from 6% to 3% by mid 1972. The 1974-75 recession caused inflation to fall from 12% to 5% by 1977.

    All these recessions were caused by steep increases in interest rates by the Fed to tackle increases in inflation.

    Now one could relate the rising and falling of the natural interest rate with the rising and falling of inflation expectations. But one can make the same correlation between the rising and falling influence of labor. The year 1975 appears to be the tipping point for labor: when union membership starts dropping and real median male income stops growing along with real GDP-per-capita. Of course, it took decades of disinflationary monetary policy to neuter labor.

    The reason the labor hypothesis is stronger than rational expectations: oil shocks in 1973 and 1979 caused extremely high spikes in inflation (13% & 15%), but similar oil shocks in 2003-2008 and 2011 had a much smaller effect on inflation.

    The 2008-09 recession caused inflation to drop from 5.5% to -2%. Although not triggered by the Fed, it is having similar labor-busting effects, which is causing inflation to come in lower than the (artificial) 2% target. It's also preventing a real recovery.

    The US unemployment rate is a poor indicator of labor recovery. The male employment rate aged 25-54 is a better barometer. It has been steadily falling since the 1974-75 recession: employment peaks corresponding with unemployment troughs: 1973, 92.6%; 1979, 91.0%; 1989, 89.9%; 2000, 89.0%; 2007, 87.5%; latest: 2013, 82.8%.

  11. "It may also be because I’m old enough to remember the 1970s when there were still economists around ... who thought any relationship could be negated by direct controls on wages and prices, with disastrous results."

    Monetary policy over the past 35 years ended in far more disastrous results (dot com bubble + housing/financial bubble + meltdown) than the high inflation of the 1970s.

    But the solution to 1970s inflation is the same as depression economics: that is, using government policy (fiscal, monetary and regulatory) as a counterweight to the business cycle. In depressions and bouts of high inflation the counterweight has to swing far.

    So a more effective way than using monetary policy alone — to either reign in high inflation or stimulate aggregate demand in a slump — is to add fiscal and regulatory policy. One has to take off the ideological lenses to imagine how that might work.

    In the 1970s, wage and price controls weren't successful (price controls are impossible because there are too many goods and services to keep track of.) But if wages were frozen that would've put a stop to corresponding price increases and the huge spikes in inflation that peaked in 1975 and 1980. Taxes could have been raised to dampen economic activity and reign in inflation. In short, cooperative inflation targeting is a much more effective way to manage inflation (with the least amount of government debt.)

    Economists also need to look at other ways of raising living standards than with rising wages (which can be inflationary.) As GDP increases, public benefits can be increased with a limited effect on inflation. A counter-cyclical variable-rate VAT could be a more direct way to manage prices than monetary policy (especially if people are aware of how the policy works.)

    BTW, notice how Reagan's 1982 income tax cuts (fiscal stimulus) made the Fed's job worse? Raising income taxes would've been the smarter move.

  12. Simon, you asked at the top of the post. "Suppose you had just an hour to teach the basics of macroeconomics, what relationship would you be sure to include?"

    I would start with the following:-

    The primary economic identity of a sovereign floating FIAT currency economy is:

    (S – I) = (G – T) + (X – M) .

    The three sectoral balances (the terms in brackets) have to sum to zero at all times. (I paraphrase Bill Mitchell).

    The capital letters refer to: Total private saving (S), which is the difference between total private sector spending and disposable income. Total private investment (I). Total government spending (G). Total government taxation revenue (T). Net exports (X – M), where X is exports and M is imports.

    The sectoral balances derived are:

    The private domestic balance (S – I), is positive if in surplus, negative if in deficit. A deficit means the private domestic sector is spending more than its income and vice-versa.

    The Budget balance (G – T), is positive if in deficit, negative if in surplus (as written). A deficit means the “government”, (the Treasury and its Central Bank), THE CURRENCY ISSUER, is adding net financial assets to the non-government private sector THE CURRENCY USER.

    The Current Account balance (X – M), is positive if in surplus, negative if in deficit. A deficit means that the external sector is draining domestic demand (spending). That is, the rest of the world is amassing buckets full of our FIAT currency, selling us mobile phones and 60 inch plasma TVs.

    For the left-hand side of the equation to be positive, (that is, the private domestic sector is running a surplus overall and net saving), then the sum of the government budget deficit and external balances, must be positive (and equal to the left-hand side).

    Should the external sector be in deficit (X less than M) by say, 3 per cent of GDP, then if the private domestic sector is to save overall, then the government sector would have to be in deficit (G more than T) of, at least, 3 per cent of GDP.

    So you immediately see what is wrong with imposing a balanced budget in an economy that uses a FIAT currency, that is not convertible into anything but more FIAT currency. The current government is using a monetary system which became redundant, when we came off the gold standard with its CONVERTIBLE currencies.

    Politicians have to realize (or simply admit) we are not on the gold standard any more. We run FIAT currency economies now, which allow extensive fiscal control that monetary control just can't achieve. Government Debt and Deficits matter on a gold standard with convertible currencies; they don't in floating fiat currency economies. Politicians would have to find another major myth to peddle for votes, if the populace understood that fact.

    Having your own sovereign FIAT currency that floats in a market with other FIAT currencies you trade with is ideal. Having to use a foreign currency you have no control over, like in the Eurozone, should be avoided. You become a currency USER in those circumstances, instead of a currency ISSUER.

  13. I agree with you, the Philips curve is a key point if not the most important,
    thought we (a gov. alligned forecasting firm) also try do muddy waters when we discuss about the Philips and most of the time try to avoid it alltogether.
    However there maybe a simple explanation of why "it is difficult to find consistency with the data" and it does not even require microfounding (sorry for that).

    Simply we might be looking into a two-dimentional projection of a multi-dimentional phenomena.
    If you consider different periods independently (e.g. 75-84, 85-94,95-present) you can find well defined correlations.
    What is changing during these periods that we are ignoring?
    maybe we don't have inflation expectation right? but that is a hard variable to measure anyway.
    If we introduce labour flexiblity provided by OCSE,
    for instance (but there may be more than that), mud clears out a lot and the Curve re-appears clearly, shifted by changing work flexibility: every 1% of increase in labour flexibility pushes the curve down 6%.

    This is difficult to accept from a neo-classical economist because it would imply that inflation is a 'political' variable (reflecting discipline inposed onto workers) rather than a technical one.


Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.