Winner of the New Statesman SPERI Prize in Political Economy 2016


Saturday 7 June 2014

How to change the inflation target

The longer interest rates stay at the Zero Lower Bound (ZLB), the stronger the case (pdf) for raising the inflation target becomes. (No, that is not a good reason to raise interest rates today, but perhaps it helps explain why some are so keen to do so!) However there seem to be two political barriers to this happening. Central banks seem to live in mortal fear that any move to raise inflation targets will shatter their ‘hard won credibility’ and inevitably lead to inflation ‘taking off’ and inflation expectations becoming ‘unhinged’. Where politicians have control (as in the UK), they worry the public will interpret any increase in the inflation target as a further erosion of their living standards. Yes, I know Abe raised the target rate in Japan, but only to the 2% that has become the consensus for the major developed economies.

Tony Yates suggests a very British solution to this problem. In the past, a standard way that UK politicians have handled such tricky questions has been to establish an independent commission to examine the question. (One famous example in the past was the Macmillan committee, set up after 1929 to establish the causes of the UK depression. Keynes’s role on that committee is vividly described by Peter Temin & David Vines in their recent book.) Tony calls it the Inflation Remit Review Commission. This commission could take evidence (from the central bank and others) on issues such as what the medium term natural real interest rate is likely to be, and compute the costs and benefits of any change. This sounds like a good way of trying (as far as possible) to depoliticise the issue, and putting the central bank’s inflation paranoia in context.

I have just one suggestion to add. This commission, at the same time, should examine whether it remains appropriate that the inflation target should just involve the consumer price index (and in particular, whether the rate of change of wages should also be targeted), and whether the target should involve the inflation rate or a path for prices (a price level target), or indeed national income. These questions naturally go together with the choice of the level of any inflation target. If there is a target for the path of prices rather than its rate of change, then the consequences of hitting the ZLB are less severe, for example. In addition, the possibility of ending the tyranny of the consumer price index, or changing to a level target, suffer from much the same conservative bias from politicians and/or central banks as the value of the target itself.

Whether the idea of a commission would work in the US I do not know. In the Eurozone there are strong grounds for setting up a similar body, not only to review the questions set out above, but also to permanently oversee the performance of the ECB. At present the ECB has minimal accountability. Appearing every three months before the European Parliament just does not count, I’m afraid. (For further discussion, see this Bruegel paper by Claeys, Hallerberg and Tschekassin.) This in itself is a strange state of affairs for an institution with such power. In addition its performance since 2011 (at least) has been very poor relative to the Fed and Bank of England. (See for example this post written over a year ago which I think reflected the clear consensus among macroeconomists at the time.)

Tony does not suggest who might sit on this commission, except to say that they should not be part of the government, or the central bank. My own view is that it is vital that at least some are academic macroeconomists, because academic macroeconomists do understand these questions better. For the UK obvious candidates are ex-external members of the Monetary Policy Committee, unless you think they have become (or were perhaps selected because they were) too indoctrinated with the central bank view. Suggestions on who might comprise an oversight commission for the ECB would be very welcome.      

10 comments:

  1. Good suggestions!
    I passed them on my contacts in the European Parliament.

    ReplyDelete
  2. " This commission, at the same time, should examine whether it remains appropriate that the inflation target should just involve the consumer price index (and in particular, whether the rate of change of wages should also be targeted), and whether the target should involve the inflation rate or a path for prices (a price level target), or indeed national income."

    Hear, hear. Given the present government's aversion to using fiscal policy to maintain demand, and since the current monetary policy arrangements were set up under Labour, one would have expected the coalition to start exploring these issues as a matter of urgency when they came to power in 2010. It's long past time that we had a serious debate on the question of monetary policy, including an evaluation as to what extent the inflation targeting regime may have contributed to the crisis, and what polices are best at coping with the ZLB.

    I wouldn't bar former MPC members from sitting on the commission, but I think it would be good to include some independent academic experts as well.

    Incidentally, I don't have a strong view on what the correct policy regime should be (perhaps it would be best to stick with the current policy combined with a commitment to use fiscal policy at the ZLB) but if this isn't the time to take stock and discuss the issues I don't know when is.

    ReplyDelete
  3. You could discuss Shiller's 'baskets' as well, automatically index-linking inflation to 'money': it'd never go through, but it would be a nice position to have as an extreme, making inflation target raise to 4% or 5% with a plus/minus 1% band easier to negotiate.

    ReplyDelete
  4. Is there no problem in this world that would not be better dealt with if given to committees of Philosopher Kings (drawn from the better universities) rather than by these oafish elected pols?

    ReplyDelete
  5. The ECB was setup as independent, because Germany insisted on it, and put it into the treaties.

    Those endless cheap attempts to introduce criminal institutions like your commision. You can never give up, do you?

    ReplyDelete
  6. I nominate Scott Sumner to head up the Commission.

    ReplyDelete
  7. On the ECB - I'm afraid you are mixing two issues: 'inflation targets' and 'centhar bank accountability'.
    On the former, we need a political process to decide the optimal (rule for resetting the) inflation target, and an expert commission to inform the political process.
    On the latter, I'm a macroeconomist myself, and I agree that SOME (international) macro academics should be on board supervising the ECB... But they should be elected by the Parliament, and there should be trade union and civil society guys on board as well. You don't need to be an economist to understand, for ex., that average inflation in the EZ since 2009 has been 1,3 i.e. way below target, nor you need to be a jurist to understand that Draghi' continuing claim that their 'only' target is price stability is an interpretation at odds with EU Treaties. There's enough tecnocracy and not enough democracy in the supervision of the ECB: improving the tecnocracy regime does not get the point. A final remark: the eu parliament has already produced criticism towards the ecb, to no effect. Unless the parliament has the power to name/remove/change centran bankers, it will lack teeths to truly supervise the bank; this is the crucial issue for an effective governance, for a well understood 'independence' of the ecb (on instruments, not on objectives); and I would welcome your comments or a post.

    ReplyDelete
  8. «whether it remains appropriate that the inflation target should just involve the consumer price index»

    "Inflation target" and "consumer price index" are often vaguely related mostly because what "inflation" means is far from uniquely determined and so is whatever is measured by the "consumer price index".

    As to "inflation target" there are at least two popular and completely different meanings of "inflation":

    * The first is the one that central banks and macro-economists love, and comes from Friedman's prescription (as opposed to description) that only "a monetary phenomenon" may be considered "inflation". Because only a generalized increase in the mythical "quantity of money" generates inflation, by increasing nearly all prices. If some prices go up and some go down it is not "inflation", even if most go up and not many go down, because it is not a change in the "absolute" level of prices so cannot be an increase in the "quantity of money" but rather a movement in relative prices "because markets". The politics are simple: the most obvious symptom of a general increase in the quantity of money is a rise in wages, because wages enter as costs in most products. Increases in profits and asset prices are never "inflation" because they are always due to increased productivity of capital, that is "relative" changes in prices.

    * The second is a definition instead of a prescription and it is based on whether the purchasing power of the same amount of currency decreases or increases *given an invariant basket of goods or services* (that is a basket including the same good types and quantities) across time. Usually the basket is presumed to be the one consumed in a month or a year by the *median family* (that is low-income, in the UK marketing slices around C2 level) at some point in time. The politics of this are simple too: measure purchasing power across time for median families, to ensure the national "company store" is delivering the same value across time, and that wage increases are not clawed back by rising prices at the "company store" and wage cuts are not delivered by stealth by the "company store" increasing prices; because what median families are interested in is the purchasing power of each dollar or pound they earn by negotiating their wages.

    ReplyDelete
  9. Also the two meanings, prescriptive and descriptive, are not only very different, but they also imply completely different sampling techniques:

    * Under the prescriptive monetarist meaning, inflation is measured without regards to *quantities* traded: one measures as many prices as possible (except profits and assets prices!), and if the "quantity of money" has inflated by 20%, then all prices (including wages...) will rise by 20% because (usually) monetarists and other "aligned" macroeconomists believe that money is a veil and only the "markets" determine relative prices as there is only one true "inflation" number, and it coincides with the the rate at which the "quantity of money" inflates.

    * Under the descriptive meaning for purchasing parity, the standard of living is defined by specific goods *and* specific quantities of those goods, which must be kept constant for inter-temporal measures of the inflation rate of that basket to be meaningful. The basket (both goods and quantities) can be revised but only rarely, says every 10-20 years, and there must be a new inflation series defined with it, and some explicit and prominent "adjustment factors" published to make different series comparable.

    The very clever trick in most countries is that despite the name, the "consumer price index", in especially the one ex-food and ex-energy, is not based on the second, descriptive meaning of "inflation", but on the first: it seems to use methodological techniques to be a proxy for the general price level, not for the purchasinThe two meanings, prescriptive and descriptive, are not only very different, but they also imply completely different sampling techniques:

    * Under the prescriptive monetarist meaning, inflation is measured without regards to *quantities* traded: one measures as many prices as possible (except profits and assets prices!), and if the "quantity of money" has inflated by 20%, then all prices (including wages...) will rise by 20% because (usually) monetarists believe that money is a veil and only the "markets" determine relative prices. Also, there is only one true "inflation" number, and it coincides with the the rate at which the "quantity of money" inflates.

    * Under the descriptive meaning for purchasing parity, the standard of living is defined by specific goods *and* specific quantities of those goods, which must be kept constant for inter-temporal measures of the inflation rate of that basket to be meaningful. The basket (both goods and quantities) can be revised but only rarely, says every 10-20 years, and there must be a new inflation series defined with it, and some explicit and prominent "adjustment factors" published to make different series comparable.

    In many countries despite its name the "consumer price index", especially the one ex-food and ex-energy, is not based on the second, descriptive meaning of "inflation", but on the first: the methodology used to build it seems aimed at creating a proxy for the general price level, not for the purchasing-parity based constant standard of living of a median family.

    ReplyDelete
  10. Sorry for the repetition above on sampling, editing went wrong.

    To compensate, a link to a page with a fascinating graph about how some countries compute CPI:

    http://signsofchaos.blogspot.com/2014/05/volatility-of-components-of-consumption.html

    The graph is the second one and it is made of three lines for the price index components for consumer service, consumer non-durable, and consumer durables. The first two grow much at the same pace for decades; the consumer durables price index line instead suddenly inverts in 1995 and since then has been going down steadily. Amazing but truthy. It probably had a a rather significant beneficial effect on "real" GDP reports since 1995.

    As Larry Summers wrote "rather energetic hedonic effort" happens:

    http://conversableeconomist.blogspot.co.uk/2014/01/larry-summers-who-always-has-something.html


    The Gerschenskron effect (and much else) lives again after the end of the Soviet Union! :-)

    ReplyDelete

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.