We need (low) inflation to prevent (high) inflation

In a recent paper, John Cochrane discusses a scenario where fiscal debts become so burdensome that the central bank is virtually forced to inflate:

At that point, inflation must result, no matter how valiantly the central bank attempts to split government liabilities between money and bonds. Long before that point, the government may choose to inflate rather than further raise distorting taxes or reduce politically important spending. Argentina has found these fiscal limits. So far, the U.S. has not, at least recently.

Argentina is certainly an instructive case, but I’m not sure it shows what Cochrane thinks it shows.  The government of Argentina adopted a highly inflationary policy in 2002, partly in an attempt to dig out from under a big debt burden.  But what sort of policies preceded that decision?

Around 1990 Argentina began some neoliberal reform, which triggered very fast growth in RGDP during 1991-97.  Then it all fell apart.  Argentina had decided to adopt a currency board as a way of tying the hands of the central bank, to prevent a repeat of the previous hyperinflation.  Big mistake, they should have followed Chile’s decision to target inflation.  In 1997-98 many developing countries got into trouble, and devalued sharply.  Argentina lost competitiveness.  Then the high tech boom caused the dollar to strengthen, even against the currencies of other developed economies.  Because Argentina was fixed to the dollar, their peso appreciated even more.  Now they were hopelessly uncompetitive, and tried to restore competitiveness under the only method allowed by a currency board, lower wages and prices, aka internal devaluation.  Argentina entered a 4 year long depression as prices fell and unemployment rose to over 20%.  The GOP made the exact same mistake in the early 1930s.

With NGDP falling so sharply below trend, the debts became politically impossible to manage.  Just as in the US during the 1930s, a new left wing government came in, abandoned the fixed exchange rate and defaulted on debts (via both inflation, and a reneging on the gold/dollar clause in contracts.)

And in both cases RGDP started growing really fast, despite horrible statist polices 10 times worse than anything Obama could imagine in his wildest dreams.  So much for RBC theory.

Conservatives were to blame for both crises.  So if conservatives want to prevent high inflation (and I certainly do) the best way is to make sure NGDP grows at a steady rate.  We stopped doing that in 2008, and we are now paying the price.  A little inflation now may prevent a lot more inflation later.

At the University of Chicago they are skeptical that wages could be sticky for a long enough period to explain the current unemployment rate.  The new classical economists generally do accept that demand shocks can have real effects in the short run, but think wages should adjust within a year or two.  I certainly would agree that not all unemployment is due to deficient AD, there was also the big jump in the minimum wage and the extended UI benefits.  But I think much more of it is AD-related than most people realize.  Paul Krugman has an important recent post that shows just how sticky wages become near zero percent inflation.  There is a shocking discontinuity in the distribution of nominal hourly wage gains at 0%, a pattern that is not consistent with New Classical models.  Those models work well at relatively high inflation rates, when all you need to do is negotiate new contracts with smaller pay increases, but the adjustment process seems to take longer when many workers need nominal pay cuts, at least if we aren’t willing to provide a bit more NGDP.

PS.  Some people wonder about the title of my blog.  Krugman’s post contains the single most perfect illustration of money illusion that I have ever seen—the distribution of nominal wage rate changes.

PPS.  Krugman has another post on Italy, which suggests why Cochrane’s theory should not be dismissed.  I don’t think his model tells us much about the US.  But it might eventually prove to say a lot about the European periphery.



21 Responses to “We need (low) inflation to prevent (high) inflation”

  1. Gravatar of Lorenzo from Oz Lorenzo from Oz
    11. July 2011 at 18:09

    This is where I have a problem with the notion of money “illusion”. I agree, Krugman’s post is great. But why is there any illusion here? People have existing obligations denominated in money, the medium of account. They are going to have to pay them regardless of how the value of money in terms of goods and services is moving. So resisting being paid less money is perfectly rational.

    I get that the simplifying role of money means there can be “friction” in the use of money (i.e. prices in terms of goods and services can shift in ways which lead to a lag in people’s grasp of the current value of money in terms of goods and service). So, that might be called an “illusion”. But it seems more like information delay than some departure from rationality.

    Similarly contracts are denominated in the medium of account, not in what the medium account can buy in terms of goods and services. In the terms of how contracts are framed, reducing what one side receives without reducing what they provide is an offense against fairness and reliability: values basic to the existence of contracts at all. Again, why is this at all usefully called ‘illusion’?

  2. Gravatar of Benjamin Cole Benjamin Cole
    11. July 2011 at 18:13

    I think they have been trying for 20 years in Japan to bring prices down to the point where they could get great growth with no wage or price inflation.

    The results have been a much higher yen, and 20 years of deflation in equities and real estate values, and torpid economic growth, and some deflation.

    Maybe after 40 t0 60 years, tight money will work. Let me know. Or my son.

  3. Gravatar of Rien Huizer Rien Huizer
    12. July 2011 at 01:26

    I constantly wonder why the rating agencies make such a fuss about the Greek etc situation and not the situation in the US. They should simply stop rating sovereign debt and investment rules should require institutions to do their own analysis, or stay out. That would at least break this cycle of self fulfilling default suggestions. I guess many European holders (as owner or secured creditor) of public debt (the ECB included) would prefer to have no interference at this stage.

    We are in a situation that may be different from a normal business cycle, in the sense that there is less mean reversion and that the future path of GDP (R or N) in the established OECD countries is simply flatter than we are used to, in fact maybe level on a per capita basis, while the rest of the world grows, but no longer with the OECD area as the primary source of global demand. This is pretty well accepted by many Europeans and Japanese, but for the hardest hit of the OECD countries (the US) it may take a while until the public sees the light.

    Given that there is no academic consensus on what can be done to lift NDGP and certainly no altruism among politicians that could make solutions effective, it is highly unlikely that we will be able to leave this mess and also that it will be necessary to use innovation in order to deal with the current pace of debt allocation. An OECD-wide sovereign default might be the right thing, combined with subsidies for pension funds and nationalization of insolvent banks. And there may be other geniuses with even better ideas.

  4. Gravatar of James in London James in London
    12. July 2011 at 03:19

    Since Krugman doesn’t comment on comments, this comment on his blog was pretty apt: “Don’t wages generally fall by the process of higher-paying employers shutting down and being replaced by lower-paying employers? This would be downward pressure on groups of workers, not individuals. Employers also may close if they overspent on facilities or executive pay. Pay and facilities aren’t the only factors affecting who’s less competitive, of course.”

    The point being that the real economy is more complex and adjusts in many more subtle ways than the macro-AD/AS-model allows.

  5. Gravatar of Gregor Bush Gregor Bush
    12. July 2011 at 05:17

    “Because Argentina was fixed to the dollar, their peso appreciated even more. Now they were hopelessly uncompetitive”

    Absolutely. And don’t forget that after the 1997-1998 Asia crisis, commodity prices plunged, which made thier competetiveness problem even worse. The world market prices of thier exports were falling as thier currency (the USD) was soaring.

    The difference in the composition of the terms of trade between Argentina and the US should have made it obvious that the currency board was a bad idea.

  6. Gravatar of Lucas Lucas
    12. July 2011 at 15:00

    Cochrane is fully sold on the fiscal theory of the price level. I’m amazed at the number of top economists who believe it to be true. Here in Argentina, it’s common to hear (from prestigious and orthodox economists) that our past episodes of hyperinflation were the result of huge fiscal deficits, relegating the role of the CB to monetizing those deficits. They still think that fiscal policy plays a big role in determining the price level [1] (Spanish)

    I agree with your assessment of the Argentine crisis and you can use it as a yardstick to measure the extent of Greece’s problems:
    – In 2001, our fiscal deficit and debt/GDP ratios were much lower than Greece’s and even lower than 2011’s America.
    – After the default and devaluation, we were blessed by a big positive shock to terms of trade (the commodity boom).
    – Our demographic situation was (and is) much better than Greece’s. Greece has an old and shrinking population [2]
    – We didn’t have to build a new monetary structure from scratch, which is what would happen if Greece exits the euro.
    – Our economy was far less interconnected to the world economy than Greece’s economy is, mainly through financial channels.
    I’m sorry for instilling pessimism, but Greece is truly between a rock and a hard place and the ECB is pursuing a hard money policy. That’s insane.

    1- http://colectivoeconomico.org/2011/06/17/la-politica-fiscal-en-la-macro-%C2%BFque-hay-de-nuevo-viejo/
    2- http://en.wikipedia.org/wiki/Demographics_of_Greece

  7. Gravatar of Scott Sumner Scott Sumner
    12. July 2011 at 18:18

    Lorenzo, You said;

    “This is where I have a problem with the notion of money “illusion”. I agree, Krugman’s post is great. But why is there any illusion here? People have existing obligations denominated in money, the medium of account. They are going to have to pay them regardless of how the value of money in terms of goods and services is moving. So resisting being paid less money is perfectly rational.”

    Sorry to say this, but this is money illusion. Sure, we always want more to less. But many expenses do rise with inflation. So any sharp cut-off should rise with inflation as well (not all expenses are fixed in nominal terms.) In economics there is nothing magic about zero. In real life there is. That’s money illusion, as the sharp break occurs at zero whether inflation is 2% or 3%.

    Benjamin, I agree.

    Rien, Greece doesn’t have it’s own currency. For better or worse that’s why the markets treat them so differently.

    James, That’s true, but the AS/AD model does allow for that.

    Gregor, Yes, I forgot about the commodity prices.

    Lucas, Those are all excellent points. In fairness to Cochrane, in the 1980s the deficits might have been so bad that, if the central bank wasn’t independent, it was virtually forced to monetize.

  8. Gravatar of Lucas Lucas
    12. July 2011 at 20:03

    “the deficits might have been so bad that, if the central bank wasn’t independent, it was virtually forced to monetize.”
    Then, the fiscal theory of the price level is a special case of the quantity theory. It’s true when the CB, because of political/institutional failings, isn’t independent enough.

    Greece scares me. There are reports of bank runs in slow motion and political violence is on the surge. It’s Argentina in 2001. I was 16 back then and the same set of events occurred in almost the same order.
    The world truly needs Rooseveltian resolve and fear nothing but fear itself.

  9. Gravatar of James in London James in London
    12. July 2011 at 23:33

    Scott you said “the AS/AD model does allow” for dynamic change within the economy that puts downward pressure on wages. I am not so sure. Krugman’s blog comment you linked to only showed wage deals for continuing employment.

    I think the AS/AD model is so all encompassing it explains everything and, therefore, nothing. It’s as if you are saying “it’s ‘aggregate’ and therefore everything is in it. A bit like a God-theory of theory of the universe.

  10. Gravatar of James in London James in London
    12. July 2011 at 23:33

    The other day you pointed to the Lucas critique of fiscal policy, what about a Lucas critique of monetary policy? Why shouldn’t targetting NGDP have negative unintended consequences too?

  11. Gravatar of W. Peden W. Peden
    12. July 2011 at 23:47

    James in London,

    With AD/AS, I do think it’s sometimes necessary to convert it into something more practical, especially when it comes to microeconomic phenomena. After all, businesspeople think in terms of price and stock, not supply and demand.

    Also, the Lucas critique is not that there are unintended consequences (which is a truism known before the mid-1970s) but that economic models should take economic policies and forward-looking expectations into account when explaining and predicting behaviour. As I understand it, Prof. Sumner’s NGDP targeting proposal is BASED on this premise i.e. a central bank committing itself to an explicit target will have stabilising effects on the economy.

    More problematic is Goodhart’s Law, which I suppose the NGDP Futures Market is supposed to resolve. Unlike Prof. Sumner, I still think it would be useful for the central bank to set out its own forecasts, looking particularly at shifts in broad money, money wages and real income growth. At the very least, this could detect any serious distortions in the Futures Market. I don’t think that would lose any of the market purism: if we want the central bank to operate like a business in response to the needs of the economy, then like many businesses we should allow it some internal forecasting and data analysis, PROVIDED that it is always gearing policy to forecast success.

  12. Gravatar of Rien Huizer Rien Huizer
    13. July 2011 at 00:56


    Greece has the Euro as its currency. Of course it has very little control over monetary policy, not enough to repair past disastrously incompetent fiscal policy. But that is the situation many countries with “their own” currency find themselves in (and then it discourages that kind of irresponsible behaviour most of the time, agreed)..I had coercive measures in mind. Inefficient, prone to abuse, unfair etc. But maybe effective enough to salvage a political project that is being vandalized by financial markets that would not have had the regulatory space to do so only thirty years ago. That is not only EU official rhetoric but also a real possibility. For politicians, those measures can be costless..

  13. Gravatar of James in London James in London
    13. July 2011 at 07:49

    @W Peden
    I can see the benefit of NGDP targetting, but I still don’t see any discussion of -ve unintended consequences. I think that printing money and passing it to banks won’t be a good thing, or are encouraged to lend their reserves rather than buy T-Bills or pass it to the Fed.

    One example is that banking is mostly done by giant, complex, beasts, that aren’t so different, economically to the GSEs. They are still way TBTF and therefore effectively government agencies. I see no recognition of this practical, real world, micro-type issue with the beautifully simple AD/AS models of the macroeconomic types on this blogsite.

  14. Gravatar of James in London James in London
    13. July 2011 at 07:52

    Also more generally, I’d like an answer to the question of how much debt is too much debt? For households, corporates, governments and countries? Various countries in Europe are now finding out, why not the US? (And please don’t say it doesn’t matter because the US can print money.)

  15. Gravatar of Scott Sumner Scott Sumner
    13. July 2011 at 13:53

    Lucas, Yes a special case of the QTM, and I’m also worried about Greece, but don’t have strong views on what to do.

    James, There is a sense in which AS/AD explains everything, but I meant something more. It’s only useful if we can identify the sorts of things that shift each curve. And I could see the example you mention very easily shifting the AS curve. So I think it fits in nicely.

    Here’s an example of what it’s not good at. The AS/AD model can theoretically explain long run growth, but that’s not its forte. Same with long run inflation. Its forte is explaining temporary adjustments having to do with the business cycle.

    The Lucas Critique might apply anywhere, but you can’t simply assert it without a specific reason.

    Rien, I’m confused, The financial markets aren’t vandalizing Greece; Greece is vandalizing the financial markets.

    James, I agree the US has waaaaaaay too much debt, and have lots of posts encouraging more pro-saving policies, higher down payments on mortgages, etc.

    You said;

    “I think that printing money and passing it to banks”

    I’ve never advocated this. Banks need not play any role in money creation. Indeed I’ve discussed negative IOR, which would dislodge money from banks.

  16. Gravatar of James in London James in London
    14. July 2011 at 02:29

    AS/AD curves don’t exist except in macro-economists’ models, the world is too micro for macro, as Lucas sort of pointed out.

    If the US has “waaaaay” too much debt as you agree, how much is too much? (Household, Corporate and Public – in all its official and unofficial varieties.)

  17. Gravatar of ssumner ssumner
    14. July 2011 at 08:46

    James, Lucas did macro too, and I have no idea how much debt is needed; that’s for the market to decide. Right now the government subsidizes debt. They should stop.

  18. Gravatar of Rien Huizer Rien Huizer
    16. July 2011 at 01:26

    Scott you said:

    Rien, I’m confused, The financial markets aren’t vandalizing Greece; Greece is vandalizing the financial markets”

    I did not say that, I said that the markets were vandalizing a political project (which I happen to like). Before the mid 1980s (only seven years prior to the decisions that set the EU on track for monetary union) national gvts, including the UK would not have facilitated this kind of market activity (I mean a handful of traders trading very small amounts of securities and derivatives in a highly opaque market and observed by a media that see this as a reality show). Not to say that there is no merit in the financial system we created since (and if you want macroeconomic efficiency, reducing market frictions caused by gvt institutions and activities would be key; I suspect that our societies are unable to cope with the consequences as appears to be the case with the cyclical corrections of the past decade: look at all that vote buying dressed up as keynesian politics, there must be a market for luddite economics).

    I agree that the Greek problem has many aspects that make it difficult to solve, especially for the Greeks themselves) but much better market governance (a work in progress stalled for years especially by the UK) for the markets in EU gvt securities (the subject of a very good, but politicall sabotaged EU project) would have made thins a lot more palatable. The volumes traded are so small that any buy back program (with EU funds) aiming at achieving an effect similar to a write-down is considered unfeasible. Yet the media claim that “the market interest” for country XYZ is now 8 times the market rate for German Bunds. Just imagine the Arkansas or California GO munis would become a benchmark (abstract from the tax aspect for a while) for US rates due to the sudden disappearance of Federal securities (China bought the lot and withdrew from the market). I am sure that the market for that sort of securities would receive very serious and constructive attention

    Of course Greece should never have been given the opportunity to exploit governance defects in the monetary union and especially the Stability Pact. Given that the vast bulk of these Greek bonds were intially placed with residents of the country itself. But that is history

  19. Gravatar of ssumner ssumner
    16. July 2011 at 10:17

    Rien, I still think you are blaming the messenger. Greece and the others are hurting the bondholders, not the other way around. They behaved recklessly, not the bondholders. (Not just Greece, but Italy, Ireland, Portugal etc.) Without liquid markets, they would be paying even higher interest rates, as people would be afraid to hold the bonds. If anything, the markets have been behind the curve, giving the countries a break they didn’t deserve. Rates have shot up recently, but probably should have been higher years ago, given the risk.

    The problem wasn’t which country bought the Greek bonds, the problem was that they never should have been sold. Greece had no business running budget deficits.

  20. Gravatar of Rien Huizer Rien Huizer
    16. July 2011 at 23:19


    I completely agree (d) that Greece should not have sold the bonds ( should not have been given the faux EUR credit card). And the parties that bought them (apart from Greek institutions obliged to do so) were reckless. The response of the EU authorities and the ECB reflects a complex of conflicting interests.

    That situation, combined with the unfinished EU process to create a proper market (not the playground that it is right now) with only a small information gap between insiders and outsiders, allows a small subsection of the market to play games that should be very much harder in good markets. That is why I believe that the default risk information coming out of this process is unreliable and should not be used for marking purposes.

    The reporting of these prices and their comments play into the hands of campaigns by political parties and media to undermine the EU (without having a clue about what the alternative ought to be) and making people believe that a Greek default would necessarily mean the end of the EUR via a variety of channels (contagion, bank insolvency, popular resistance in Germany etc).

  21. Gravatar of Scott Sumner Scott Sumner
    17. July 2011 at 16:56

    Rien, I can’t see the euro ending. Worst case Greece leaves, and maybe a few others. Countries are still joining, and more are on the way.

Leave a Reply