The Phillips Curve and the Lucas Critique

With unemployment at the lowest levels since the start of the millennium (initial unemployment claims in February were the lowest since 1973!), lots of people are starting to wonder if we might be headed for a pick-up in the rate of inflation, which has been averaging well under 2% a year since the financial crisis of September 2008 ushered in the Little Depression of 2008-09 and beyond. The Fed has already signaled its intention to continue raising interest rates even though inflation remains well anchored at rates below the Fed’s 2% target. And among Fed watchers and Fed cognoscenti, the only question being asked is not whether the Fed will raise its Fed Funds rate target, but how frequent those (presumably) quarter-point increments will be.

The prevailing view seems to be that the thought process of the Federal Open Market Committee (FOMC) in raising interest rates — even before there is any real evidence of an increase in an inflation rate that is still below the Fed’s 2% target — is that a preemptive strike is required to prevent inflation from accelerating and rising above what has become an inflation ceiling — not an inflation target — of 2%.

Why does the Fed believe that inflation is going to rise? That’s what the econoblogosphere has, of late, been trying to figure out. And the consensus seems to be that the FOMC is basing its assessment that the risk that inflation will break the 2% ceiling that it has implicitly adopted has become unacceptably high. That risk assessment is based on some sort of analysis in which it is inferred from the Phillips Curve that, with unemployment nearing historically low levels, rising inflation has become dangerously likely. And so the next question is: why is the FOMC fretting about the Phillips Curve?

In a blog post earlier this week, David Andolfatto of the St. Louis Federal Reserve Bank, tried to spell out in some detail the kind of reasoning that lay behind the FOMC decision to actively tighten the stance of monetary policy to avoid any increase in inflation. At the same time, Andolfatto expressed his own view, that the rate of inflation is not determined by the rate of unemployment, but by the stance of monetary policy.

Andolfatto’s avowal of monetarist faith in the purely monetary forces that govern the rate of inflation elicited a rejoinder from Paul Krugman expressing considerable annoyance at Andolfatto’s monetarism.

Here are three questions about inflation, unemployment, and Fed policy. Some people may imagine that they’re the same question, but they definitely aren’t:

  1. Does the Fed know how low the unemployment rate can go?
  2. Should the Fed be tightening now, even though inflation is still low?
  3. Is there any relationship between unemployment and inflation?

It seems obvious to me that the answer to (1) is no. We’re currently well above historical estimates of full employment, and inflation remains subdued. Could unemployment fall to 3.5% without accelerating inflation? Honestly, we don’t know.


I would also argue that the Fed is making a mistake by tightening now, for several reasons. One is that we really don’t know how low U can go, and won’t find out if we don’t give it a chance. Another is that the costs of getting it wrong are asymmetric: waiting too long to tighten might be awkward, but tightening too soon increases the risks of falling back into a liquidity trap. Finally, there are very good reasons to believe that the Fed’s 2 percent inflation target is too low; certainly the belief that it was high enough to make the zero lower bound irrelevant has been massively falsified by experience.

Agreed, but the better approach would be to target the price level, or even better nominal GDP, so that short-term undershooting of the inflation target would provide increased leeway to allow inflation to overshoot the inflation target without undermining the credibility of the commitment to price stability.

But should we drop the whole notion that unemployment has anything to do with inflation? Via FTAlphaville, I see that David Andolfatto is at it again, asserting that there’s something weird about asserting an unemployment-inflation link, and that inflation is driven by an imbalance between money supply and money demand.

But one can fully accept that inflation is driven by an excess supply of money without denying that there is a link between inflation and unemployment. In the normal course of events an excess supply of money may lead to increased spending as people attempt to exchange their excess cash balances for real goods and services. The increased spending can induce additional output and additional employment along with rising prices. The reverse happens when there is an excess demand for cash balances and people attempt to build up their cash holdings by cutting back their spending, reducing output. So the inflation unemployment relationship results from the effects induced by a particular causal circumstance. Nor does that mean that an imbalance in the supply of money is the only cause of inflation or price level changes.

Inflation can also result from nothing more than the anticipation of inflation. Expected inflation can also affect output and employment, so inflation and unemployment are related not only by both being affected by excess supply of (demand for) money, but by both being affect by expected inflation.

Even if you think that inflation is fundamentally a monetary phenomenon (which you shouldn’t, as I’ll explain in a minute), wage- and price-setters don’t care about money demand; they care about their own ability or lack thereof to charge more, which has to – has to – involve the amount of slack in the economy. As Karl Smith pointed out a decade ago, the doctrine of immaculate inflation, in which money translates directly into inflation – a doctrine that was invoked to predict inflationary consequences from Fed easing despite a depressed economy – makes no sense.

There’s no reason for anyone to care about overall money demand in this scenario. Price setters respond to the perceived change in the rate of spending induced by an excess supply of money. (I note parenthetically, that I am referring now to an excess supply of base money, not to an excess supply of bank-created money, which, unlike base money, is not a hot potato that cannot be withdrawn from circulation in response to market incentives.) Now some price setters may actually use macroeconomic information to forecast price movements, but recognizing that channel would take us into the realm of an expectations-theory of inflation, not the strict monetary theory of inflation that Krugman is criticizing.

And the claim that there’s weak or no evidence of a link between unemployment and inflation is sustainable only if you insist on restricting yourself to recent U.S. data. Take a longer and broader view, and the evidence is obvious.

Consider, for example, the case of Spain. Inflation in Spain is definitely not driven by monetary factors, since Spain hasn’t even had its own money since it joined the euro. Nonetheless, there have been big moves in both Spanish inflation and Spanish unemployment:

That period of low unemployment, by Spanish standards, was the result of huge inflows of capital, fueling a real estate bubble. Then came the sudden stop after the Greek crisis, which sent unemployment soaring.

Meanwhile, the pre-crisis era was marked by relatively high inflation, well above the euro-area average; the post-crisis era by near-zero inflation, below the rest of the euro area, allowing Spain to achieve (at immense cost) an “internal devaluation” that has driven an export-led recovery.

So, do you really want to claim that the swings in inflation had nothing to do with the swings in unemployment? Really, really?

No one claims – at least no one who believes in a monetary theory of inflation — should claim that swings in inflation and unemployment are unrelated, but to acknowledge the relationship between inflation and unemployment does not entail acceptance of the proposition that unemployment is a causal determinant of inflation.

But if you concede that unemployment had a lot to do with Spanish inflation and disinflation, you’ve already conceded the basic logic of the Phillips curve. You may say, with considerable justification, that U.S. data are too noisy to have any confidence in particular estimates of that curve. But denying that it makes sense to talk about unemployment driving inflation is foolish.

No it’s not foolish, because the relationship between inflation and unemployment is not a causal relationship; it’s a coincidental relationship. The level of employment depends on many things and some of the things that employment depends on also affect inflation. That doesn’t mean that employment causally affects inflation.

When I read Krugman’s post and the Andalfatto post that provoked Krugman, it occurred to me that the way to summarize all of this is to say that unemployment and inflation are determined by a variety of deep structural (causal) relationships. The Phillips Curve, although it was once fashionable to refer to it as the missing equation in the Keynesian model, is not a structural relationship; it is a reduced form. The negative relationship between unemployment and inflation that is found by empirical studies does not tell us that high unemployment reduces inflation, any more than a positive empirical relationship between the price of a commodity and the quantity sold would tell you that the demand curve for that product is positively sloped.

It may be interesting to know that there is a negative empirical relationship between inflation and unemployment, but we can’t rely on that relationship in making macroeconomic policy. I am not a big admirer of the Lucas Critique for reasons that I have discussed in other posts (e.g., here and here). But, the Lucas Critique, a rather trivial result that was widely understood even before Lucas took ownership of the idea, does at least warn us not to confuse a reduced form with a causal relationship.


21 Responses to “The Phillips Curve and the Lucas Critique”

  1. 2 Henry Rech March 30, 2018 at 3:15 pm


    If there is a demonstrable correlation between the level of employment and inflation, how would you rationalize this relationship? What are the causal paths and links between inflation and employment as you see it?

    Liked by 1 person

  2. 3 Rob Rawlings March 30, 2018 at 6:02 pm

    I think the correct link to Krugman’s recent post is

    The one supplied above is to a 2013 post also (unfairly in my view) criticizing David Andolfatto.

    Liked by 1 person

  3. 4 Benjamin Cole March 30, 2018 at 9:01 pm

    Great post.

    Send in the choppers, and don’t stop until you hit 3% inflation/

    Liked by 1 person

  4. 5 EugenR March 31, 2018 at 2:57 am

    Prices ceased to be long time ago connected to limited available labour force. Prices are driven by scarcity or abundance of product or one of its components, without alternative. It is truth in micro level of individual product, as well as at aggregate macro level.

    If the 1973 stagflation didn’t give enough empirical evidence that Philips curve doesn’t work, the 2018 economic situation should. There is no correlation in the technology based economy between prices and unemployment. Prices in technology driven economy, with automatization and free information, available to the suppliers as well as to the consumers, have the propensity to reach marginal cost, which by itself have tendency to become zero or close to zero, after the initial investment was done, and the technology became available to all.

    In an economy with full automatization, mobility and information, with relatively abundant capital for investments, only new inventions and innovations, that produce temporary state of monopoly, can still generate profit, before the new invention will become a general public knowledge, and allow price well above marginal cost of the product. This temporary state could exist for some time, even decades in the pre-information economy, but not anymore. In information based economy, price increase well above the basic margin cost, can happen only if scarcity prevails in products or in inputs needed to produce it. Other way to increase price above marginal cost is by creating legal obstacle for usage of new technologies, or of unique brand. By the way, the legal protection of knowledge or copyright is becoming less and less obvious and maintainable. Viz. the situation in pharmaceutical industry or entertainment industry.

    The monetary reasoning for inflation, in technologically driven, global economy doesn’t work on ordinary consumption products, unless there is scarcity of capital capacity or substantial raw material or component needed to produce it. Since all the products are energy driven, the cost of energy is crucial. At 1973 oil had no alternative, so its scarcity caused huge inflation and also unemployment. If the scarcity is of final product, its price will increase and it will become a local event, but if the scarcity is of raw material, so basic as energy producing raw material, it causes bottleneck in production and as result of it, in one hand it will cause scarcity and price increase in large range of products and on other hand unemployment. There are very few raw material items in contemporary information and technology driven market and capital economy, without technological alternatives. Even in energy industry, the non conventional solutions, have only capital limitations, and no resource limitations.

    The result of all this is low inflation in the ordinary consumption products, except those, dependent on scarce components, like land, weather, water, environment. The next major price increase will be caused by one of these items. Just to make it clear, agriculture, even if land dependent, has still long way to utilize technologies in the food production processes, and there are many alternatives to classical Iand dependent products.

    Does it mean that monetary easing has no influence on prices? Definitely not. Money, is in one of its aspects a raw material, essential to manage economy. As such, if in scarcity its price increases and if abundant it’s price decreases. And there is other issue with money, very uncommon for other form of commodities. Money has unlimited demand because of its attribute of value holding property. Yet money as value holding item has alternatives, like investments in real estate or any other income generating assets. Here the price of money plays an important function. If the price of money, the interest, is low, what means from money holders point of view a relatively high price for money holding, the alternative investments in value holding assets seems to be more attractive. But since all the alternative value holding assets, be it real estate or company shares are limited and final at the certain time period, their prices are wildly escalating at times of low interest rate and plummeting at times of high interest rate.

    Liked by 1 person

  5. 6 Rob Rawlings March 31, 2018 at 7:09 am

    Krugman uses the example of Spain where (he claims) an inflation rate lower than its euro-zone partners led to lower relative costs and increased demand for its goods which led to lower unemployment. I think the same thing could apply to countries on the gold standard.

    In these cases of countries sharing a currency there does seem to be a case for saying that there is a causal relationship between (relative) inflation and unemployment (albeit one in the opposite direction to the one predicted by the Phillips curve).

    Liked by 1 person

  6. 7 Miguel Navascués March 31, 2018 at 11:58 am

    Great post, David. But I’m not convinced that Mr Phillips was saying that unemployment is the direct cause of more or less inflation. As you say, it’s a coincidan relationship. In my opinion it is the result of the imperfection of markets, that make impossible what Krugman call “the inmaculate inflation”. That is, is a refutation of the Lucas’ Critique.

    Liked by 2 people

  7. 8 BQ March 31, 2018 at 12:28 pm

    The paradox to this is that high employment will inevitably increase inflation. More purchasing power through well-paid workers who will in turn, lead individuals to demand more and better goods and services – even if it has to be imported. Higher imports however, will skew the trade deficit as higher imports than exports will affect the nation’s currency. Countries like the US and and UK are obsessed with the value of their currency and will do ‘what is required’ to keep it ‘strong’.

    Trump is mad at Germany because thought relatively strong and a wealthy country (with the abiltiy to bail out other European countris to promote the strength of the Euro) and with a low unemployment rate, they import very little. This keeps the currency strong. Germany’s unemployment rate has only changed since it strategically let in more migrants through its borders.

    When the obsession with strength currency by governments is alleviated, we can see less of need to bother with inflation and tackle the real issue of employment – or at least a way to empower people to make a living to live a decent life. The cost of living (inflation) can also be reviewed internally by each country through a revision if its own PPP and CPI weightings. Easier said then done. no doubt it will be a lot of hard work – but I get the feeling no one is willing to get their hands dirty.


  8. 9 EugenR April 2, 2018 at 2:18 am

    Your ideas about trade deficit and currency are looking left overs of times of national and marcentile economies, where commerce and capital flow is limited and restrained. (historically the last marcantilism was the British colonial empire). This times are long time over. The global markets and global capital took over the lead, (unless some crazy president will be successful with turning back the time). In global economy the capital flows, where primarily it promises highest political and legal security and stability, and secondarily it promises the highest yields. This is why US can maintain its huge trade deficit already three decades, and China has outflow of capital. This may change if the position of US dollar will change. There are several threats on the horizon, that may endanger the US dollar position.
    A. The first and imediate threat is making the Chinese RMB a freely convertible currency. On one hand China measured in merchandise production performance, has become the biggest economy in the world, it always will be an underdog to US or Euro based economies. What gives strength to these economies is not their production capacity, that is still there, even if not fully used, but their financial system, supporting their markets.
    B. The cryptocurrencies. Even if still marginal, and all the governments and financial institutions try to keep it marginalized, the idea of blockchain and smart contract technology can become a leading intermediation tool for value exchange. I can imagine an economy, based on idea of barter, where no intermediating money is involved, and products are exchanged as against other products, for contracted exchange value, without money as intermediation. Of course it would mean catastrophy to the existing

    Liked by 1 person

  9. 10 BQ April 2, 2018 at 2:27 am

    Economy based on barter is a true nature of trade…there would still need something as an intermediate


  10. 11 EugenR April 2, 2018 at 2:30 am

    My comment slipped unfinished, i will answer to all in my blog very soon.


  11. 12 BQ April 2, 2018 at 2:45 am

    …sorry didn’t finish writing, phone died 🙂

    an intermediary currency, but not just one (perhaps Cryptocurrency is the key and a possible fix especially because it isn’t and hopefully backed by any single government, institution or political agenda via its transparent and decentralized structure). Of course it is not the singular solution but the most neutral …Choosing a single currency (tied to one nation) is paramount to a mercantile-like mode of trade which you correctly highlighted as something of the past (and belongs to the old days of colonies).

    Maintaining a huge deficit is not healthy and relies on artificial mechanisms to keep it going until an inevitable burst. Re-evaluate what constitutes and contributes to domestic inflation and you can solve the problem rather that using and manipulating external factors (patchwork) to resolve an endemic disease.

    Liked by 1 person

  12. 14 David Glasner April 2, 2018 at 6:49 pm

    Nick, Thanks.

    Henry, The original Phillips Curve was a plot of points representing combinations of the rate of unemployment and the rate of increase in wages published in an article in the late 1950s by a distinguished economist at the London School of Economics, A. W. Phillips. The curve was downward sloping. Other economists found similar correlations between price inflation and unemployment. My point in the post is that there is very little reason to believe that there is a strong causal relationship between inflation and unemployment. Rather both variables respond to shifts in aggregate demand or aggregate supply. Increases in aggregate demand tend to raise prices and employment, decreases in aggregate demand have opposite effects. Increases in aggregate supply tend to reduce prices (or inflation) and increase employment. So the observed empirical relationship depends on whether aggregate demand shifts or aggregate supply shifts predominate. That’s a very simplified account of how they are related.

    Rob, Thanks for catching that. I have fixed the link.

    Benjamin, Well we should at least aim to get back to the price level path consistent with 2% inflation over the long-term.

    EugenR, It is a category mistake to assume that the price level is determined in the same way as individual money prices. You are correct that many new knowledge-intensive products have low marginal cost and thus cannot be profitable unless the producer has some way to limit access to the product or the technology. That has implications for how our economy is operating, but not necessarily for the monetary system.

    Rob, When different countries share the same currency. Countries with rapid productivity growth will enjoy increasing real wages which will translate into rising tradable prices while countries with low productivity growth will have falling tradable prices. If nominal wages are sticky downward, the countries with falling prices will be the ones having rising unemployment.

    Miguel, I’m not sure what Phillips himself believed. I think he was just interested in the statistical correlation and did not offer much in the way of theory. I agree there is no reason to think that in the real world inflation is immaculate, though there may be circumstances in which it can be.

    BQ, Your reasoning may be plausible, but it is far from necessary.


  13. 15 Joel Bernard April 4, 2018 at 6:26 pm

    Solow (1979) recalling Samuelson and Solow (1960) discerning VISUALLY a relationship between the Phillips Curve and ALL-PRICE inflation:

    ‘I remember that Paul Samuelson asked me when we were looking at those diagrams for the first time, ‘Does that look like a reversible relation to you?’ What he meant was ‘Do you really think the economy can move back and forth along a curve like that?’ And I answered ‘Yeah, I’m inclined to believe it’ and Paul said ‘Me too.’ And thereby hangs a tale.’

    Phillips’ original curve correlated unemployment and WAGE inflation only (except for wartime). Simple supply and Demand. He also assumed that workers would get the benefit of productivity increases. They haven’t…


  14. 16 Gerard MacDonell April 7, 2018 at 8:39 am

    A very mundane point relative to the more important issues discussed in the blog post and comments, but I am struck by the popularity of the claim that underlying inflation is still “well below” the Fed’s objective.

    If the screen consensus guess for the March data is right, and if there are no revisions, then core PCE inflation is already 1.9% on a vs year ago basis.

    We can argue about whether underlying inflation is in line with the core and whether the target should be raised and whether the Phillips Curve is valid etc.

    But there is no need to wait for the April reports to confirm that the base effect dropped out in March. Measured with the precision of which mere mortals are capable, core inflation appears already to be at target.

    Incidentally, I think the Fed is taking the advice of the doves and preparing to allow inflation to run a little above 2% in the late cycle. I think they think that is required to hit their target of 2% on average, even ignoring bygones and looking strictly forward. (Inflation has recently had a tendency to get stuck low during recession and early recovery.) I realize this may seem small beer relative to raising the target.


  15. 17 Rob Rawlings April 9, 2018 at 7:08 pm

    Regarding: ‘Rob, When different countries share the same currency. Countries with rapid productivity growth will enjoy increasing real wages which will translate into rising tradable prices while countries with low productivity growth will have falling tradable prices. If nominal wages are sticky downward, the countries with falling prices will be the ones having rising unemployment.’

    I have 2 points in response:
    – With productivity growth you are producing more stuff with the same inputs so you could have both rising real wages and falling real prices.
    – I think Krugman’s point is that for countries like Spain (which have a price level in the common currency that is too high for their productivity level) once the stickiness is eventually overcome and real wages and other prices start falling then employment will increase. And (at a certain level of abstraction) it is the falling price level that drives the increase in employment.

    This has nothing to do with the Phillips curve but nevertheless it seems correct to link inflation and employment in these circumstances.


  1. 1 Links for 04-02-18 | Me Stock Broker Trackback on April 2, 2018 at 11:29 am
  2. 2 Links for 04-02-18 – Courtier en Bourse Trackback on April 2, 2018 at 11:31 am
  3. 3 Új bejegyzések – 2018. április 3. | Magyar Tudományos Akadémia Közgazdaság- és Regionális Tudományi Kutatóközpont Közgazdaság-tudományi Intézete Trackback on April 3, 2018 at 5:12 am
  4. 4 The loose tights between inflation and unemployment » OPUSNET: Pioneering the Circular Economy Trackback on May 21, 2019 at 6:07 am

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

About Me

David Glasner
Washington, DC

I am an economist in the Washington DC area. My research and writing has been mostly on monetary economics and policy and the history of economics. In my book Free Banking and Monetary Reform, I argued for a non-Monetarist non-Keynesian approach to monetary policy, based on a theory of a competitive supply of money. Over the years, I have become increasingly impressed by the similarities between my approach and that of R. G. Hawtrey and hope to bring Hawtrey’s unduly neglected contributions to the attention of a wider audience.

My new book Studies in the History of Monetary Theory: Controversies and Clarifications has been published by Palgrave Macmillan

Follow me on Twitter @david_glasner


Enter your email address to follow this blog and receive notifications of new posts by email.

Join 3,263 other subscribers
Follow Uneasy Money on

%d bloggers like this: