Clive Crook advocates a 5% NGDP growth target

Here’s Clive Crook of the FT:

But what more can monetary policy do now? Plenty. Look at it this way. Can a central bank engineer high inflation if it chooses to? Yes, always. If it prints enough money – they call it “quantitative easing” nowadays – it can cause inflation. But if it can always cause inflation, it can always stimulate demand: the second is a necessary condition for the first.

Admittedly, the limits to the Fed’s efforts to stimulate the economy are partly prudential. At the recent meeting of its policy committee, dissenters questioned whether it was right to promise explicitly, as the central bank has, two more years of very low interest rates. Inflation hawks resist the idea of further QE. Here is the central point, however: this is a disagreement about whether further stimulus would be wise, not whether it is possible.

In my view, it is both possible and necessary. The recent revisions to the figures for growth make the economic argument so strong that I wonder if politics is not influencing the dissenters. The problem is that the Fed has to explain itself, both to Congress and to the public at large. Conditions demand what critics would call an “inflationary” monetary stimulus. The Fed’s vague mandate, which calls for both price stability and full employment, is not much help. It is a fight the Fed would rather avoid.

To make the case for new stimulus, the Fed needs better arguments. The past few weeks have settled, to my satisfaction at least, a long-running debate on this very topic. Rather than targeting inflation, central banks should keep nominal incomes growing on a pre-announced path: say 5 per cent a year. Nominal gross domestic product is the sum of inflation and growth in real output – and is the variable that monetary stimulus directly drives.

Samuel Brittan made the case for this approach decades ago on this page. The crucial point – how an increase in nominal GDP breaks down between output and inflation – is not something the Fed can determine, or should have to explain. There are pros and cons to this approach, but that is the decisive political virtue of casting the target this way.

When nominal GDP falls below track, monetary stimulus pushes it back. If inflation rises temporarily during catch-up, that is tolerated. In current conditions, this makes all the difference. The new GDP figures showed demand has fallen much further below trend than had been appreciated. With a nominal GDP target, that announcement would have led investors to expect new monetary stimulus. With the implicit inflation target that the Fed is assumed to use, it did not.

Interestingly, unlike the Fed, the Bank of England has an explicit inflation target – one it has missed so conscientiously of late that many observers believe it is following an unannounced nominal GDP target. If so, it is to be congratulated, and one day its operating mandate should be adjusted accordingly.

The Fed should move in the same direction – not, obviously, at the direction of Congress, which has its hands full getting fiscal policy wrong – but at its own initiative. Exploit that bounded independence a little more. Move to a nominal GDP regime and let the markets know, in Fed-speak, that this is what you are about. You already keep telling people, quite rightly, that monetary stimulus is not and never can be a spent force. Now would be a good time to prove it.

Exactly.  I’d add that if the British really do believe inflation is too high, then ipso facto they think demand is too high.  And this means that the policy of expansionary fiscal contraction has succeeded.  Of course it hasn’t, but not because of fiscal contraction, rather because the Bank of England was pressured to “do something” about inflation.  If the BOE is fighting inflation then nominal growth will be disappointing, regardless of what fiscal policymakers do.

HT:  Marcus Nunes



23 Responses to “Clive Crook advocates a 5% NGDP growth target”

  1. Gravatar of Martin Martin
    15. August 2011 at 05:46

    Another piece of good news:

    “Here is a sentence you are unlikely to hear from any world leader: “Because the global crisis has hit hardest at the working and middle classes, who are burdened with falling home prices and stagnant incomes, we are going to make it the focus of our economic policy to restore the value of their hard-won assets and increase their wages.” Too bad – you’d think it would be a politically winning move, as well as good economic policy. But we seem to be living in a world where quack policies prevail.”

  2. Gravatar of John John
    15. August 2011 at 05:46


    How do the quasi-monetarists explain stagflation? It seems like you basically view unemployment through the traditional macro “aggregate demand” framework. However, if you look through the back of any Economist magazine, it is clear by the statistics that inflation and unemployment are at best unrelated.

    I actually think that if anything, there is a positive correlation between inflation (too much aggregate demand) and unemployment (too little) because governments stupid enough to pursue inflationary policies tend to make idiotic choices in other areas as well: labor relations, subsidies, government programs, regulation, etc. Also, unemployment, ceteris paribus, should boost inflation as there are fewer workers turning out goods and with a given money stock, fewer available goods equals higher prices.

  3. Gravatar of W. Peden W. Peden
    15. August 2011 at 06:10


    “How do the quasi-monetarists explain stagflation?”

    Simple monetarism: there is a relationship between aggregate demand and output until the natural rate of unemployment is exceeded, after which increases in aggregate demand reflect purely in increases in inflation.

    Put another way: it is a defining and influential feature of monetarism that nominal and real variables are not related at general equilibrium. So it’s very possible for a nominal variable (NGDP) to come apart from apart from real variables (real GDP or unemployment). Quasi-monetarism is as monetarist as monetarism on this score i.e. no revision to the monetarist accelerationist analysis of the Phillips Curve (it might hold in the short term but never in the long term) is proposed.

    As you say, any positive correlation between inflation and unemployment is likely a result of other stupid choices. Of course, high levels of inflation can themselves depress output e.g. due to deflationary expectations + sticky prices/wages, instability of contracts, financial instability and haywire effects on savings behaviour.

    (Another, basically tautological, cause of a positive correlation between inflation and unemployment is that, if NGDP is stable, a fall in output will cause an increase in the price level. In this case, though, it is the change in output that causes the increase in the price level rather than vice versa.)

  4. Gravatar of John John
    15. August 2011 at 06:11


    The consumer spending and incomes of the boom years were possibly because people mistakenly believed they were richer than the situation justified due to rising home values. It is impossible for politicians or central bankers to restore and/or raise specific asset values or salaries without price controls and all the problems that come with them. If they pursue a very expansionary monetary policy with this goal in mind, there is no telling what prices will rise and in what proportions. It is very important to underline that the PRIMARY BENEFICIARIES OF AN INFLATIONARY POLICY ARE THE ALREADY RELATIVELY WEALTHY PEOPLE WHO OWN LARGE AMOUNTS OF INVESTMENTS THAT WILL APPRECIATE AND THE LOSERS ARE THE LOWER AND MIDDLE CLASSES WHO DEVOTE A LARGER PORTION OF THEIR INCOMES ON DAY TO DAY PURCHASES ON ITEMS THAT WILL BE RISING IN PRICE. Inflation is the quickest way to create big divisions between the economic winners and losers.

  5. Gravatar of John John
    15. August 2011 at 06:19

    W. Peden,

    The real world, which involves no general equilibrium (and never could), clearly shows that it is possible to increase inflation (when unemployment is well below the natural rate) without making any positive impact on unemployment.

  6. Gravatar of W. Peden W. Peden
    15. August 2011 at 06:32


    That’s exactly what the monetarist model tells us: the Phillips Curve might make a short-term dent in unemployment, but it can never reduce it below its natural rate in the long run, even with constantly accelerating unemployment. The use of GE analysis clarifies that NGDP and RGDP are causally separate in the long run.

  7. Gravatar of John John
    15. August 2011 at 07:37

    W. Peden,

    If you measured NGDP in a basket of currencies or rather than dollars, I’m pretty sure NGDP would be equal to RGDP. The fact that NGDP can move without RGDP seems a consequence solely of changes in the money relation (supply and demand for money).

  8. Gravatar of Scott Sumner Scott Sumner
    15. August 2011 at 08:23

    Martin, Good find.

    John, I pay no attention to inflation, as I find it doesn’t measure anything meaningful in the economy. Quasi monetarists focus on targeting NGDP. There is strong evidence that NGDP and employment are strongly correlated. When stagflation occurs, inflation rises and RGDP falls, but that should be of no concern to monetary policymakers, as the goal should be steady NGDP growth.

    Some types of stagflation can be addressed with supply-side reforms of government policy.

    You said;

    “The consumer spending and incomes of the boom years were possibly because people mistakenly believed they were richer than the situation justified due to rising home values.”

    Consumer incomes are determined by the Fed, not confidence. Zimbabwe has had the fastest growth in incomes in recent years, and lousy confidence.

  9. Gravatar of Benjamin Cole Benjamin Cole
    15. August 2011 at 09:24

    The gold fetishists and currency worshipers have somehow cowed Ben Bernanke-san.

    An interesting question is this: Bernanke knows what he should do, but fears criticism. But he has his job until 2014. He should say “Damn the torpedoes” and do what is right. At worst, he loses his job in 2014 in a gold-nut inspired witch hunt–although more likely he will praised as the man who ended the recession.

    Even if they run him out of town, to what? To a cushy job on Wall Street with Goldman Sachs at $2 mil a year?

    What gives with Bernanke-san? I can forgive some poor middle-class fellow sans reserves but with family caving into social “pressure.” But Bernanke-san? What gives?

  10. Gravatar of bill woolsey bill woolsey
    15. August 2011 at 09:46



    a. adverse aggregate supply shocks. For example, war or anticipated war in the middle east raises oil prices, the price level rises by way of arithmetic, and a shift in the price level from a lower to a higher level is a higher inflation rate. Meanwhile, less oil available for production implies less output, and a shift from a higher to lower growth path is slower real growth. And, finally, the need to make adjustments in the allocation of labor results in higher structural unemployment.
    The quasi-monetarist view is that the least bad response to this is to keep GDP growing at a constant rate.

    b. adjustment paths to changes in the growth rate of GDP. For an acceleration, stagflation occurs as part of the adjustment from the short run to the long run. Inflation rises and the unemployment rate rises. This inflation getting worse and the unemployment rate returning to normal. For a deceleration, unemployment rises even though inflation doesn’t fall to zero. There is still inflation (though at a lower level) at the same time unemployment is rising.

    Quasi-monetarist solution: Keep GDP on a steady growth path. We don’t favor shifting to higher and lower growth rates of GDP.

  11. Gravatar of John John
    15. August 2011 at 12:05

    You’re right that the Fed was responsible for consumer incomes during the boom years as their irresponsibly loose monetary policy (specifically 2003-4) caused housing prices to really take off. People were willing and able to borrow and spend more on the assumption that the value of their house would continually increase or at least not fall. It’s called a bubble and the Chicago School should admit they exist.

    By Zimbabwe’s fast income growth are you talking more recently or a few years ago when prices were doubling every 16 hours or so? I’d imagine NGDP and nominal incomes would grow quite quickly in that type of environment, however, I don’t think many people there appreciated it as much as the economists begging for higher inflation here would seem to think.

    You say that you pay no attention to inflation, what do you think of Keynes’s saying: “”By a continuing process of inflation, government can confiscate, secretly and unobserved, an important part of the wealth of their citizens.”?

  12. Gravatar of John John
    15. August 2011 at 12:09

    Bill Woosley,

    It is impossible for the bulk of prices to rise simply because one price rises such as oil. Inflation is everywhere and always a monetary phenomenon, at least according to Milton Friedman. If the price of gas goes up, consumers and businesses have to cut back on other purchases. For the majority of prices to rise, there has to be some kind of change in the supply and demand for money. Supply shocks can possibly reduce the demand for money, as people might temporarily reduce their cash balances to pay for new goods, but that is still a monetary change.

  13. Gravatar of bill woolsey bill woolsey
    15. August 2011 at 12:46


    I didn’t say that the “bulk” of the prices would rise because an adverse supply shock.

    I said that the price level would rise because some prices rose. It is a matter of arithmetic.

    It is also true that if the quantity of money is unchanged and the real demand for money is unchanged, then the increase in the price level (by arithmetic,) will reduce the real quantity of money. The resulting excess demand for money results in reduced money expenditure until the price level falls. And so, the end result is that the price of oil is higher and all other prices are lower, leaving the price level unchanged.

    However, assuming money is a normal good, the decrease in real income that results from there being less oil to produce things results in a decrease in the real demand for money. And so, the increase in the price level as a matter of arithmetic, which reduces the real quantity of money tends to match the decrease in the real demand for money.

    The quasi-monetarist view is that keeing GDP on a stable growth path provides minimal interference with this process. And this is the least bad option. Stablizing the growth path of prices, on other the other hand, would requie a decrease in the quantity of money to match the decrease in the real demand for money due to the lower real income. With the price level higher as a matter of arithmetic, this would result in reduced money expenditures until the price level falls back to its initial value, with the price of oil higher and all other prices lower.

    I can assure you, quasi-monetarists have a very good grasp of basic monetary economics. Well, I am pretty sure I do.

    Anyway, a decrease in the supply of oil increase the price level, which involves inflation as the price level moves from a lower to higher level. It doens’t result in a continuing increase in the price level.

    But there result above is a decrease in output (or slower growth) associated with rising prices (or more rapidly rising prices.) That is stagflation.

  14. Gravatar of John John
    15. August 2011 at 13:11

    Bill Woosley

    I said bulk of prices because price level is a stupid term. There’s no such thing as a price level. That’s an important point because even in an inflationary situation, some prices may fall. The relative changes and differentials between prices are far more important during an inflationary scenario than what the CPI or some other inherently flawed index measures. These relative price changes radically affect what enterprises are profitable and to what extent. Inflation brings about a price revolution, and therefore a production revolution, instead of a single change in the imaginary price level.

    It seems that we basically agree that for a general rise in prices, the change generally has to come from a change in supply and demand for money. Your just saying that supply shocks have the ability to change this money relation correct?

  15. Gravatar of David Carnes David Carnes
    15. August 2011 at 13:16


    I’m on board with the concept of NGDP targeting, but wouldn’t a central bank actually have to target per capita NGDP? Population growth has fluctuated from 0.8%-1.4% annual rates over the post-baby boom era, which seems like a large relative difference to ignore in targeting a rate of 5% (over 10% variance). Is this a concern?

  16. Gravatar of W. Peden W. Peden
    15. August 2011 at 14:13

    bill woolsey & John,

    I’m a bit confused with your thinking, because you both seem to be talking about a pretty simple phenomenon (a fall in the price of oil, which is a fall in output) either in terms that are unfamiliar to me or in a way with which I disagree.

    Assume NGDP is growing at 10%. Assume an oil crisis reduces general production, such that deflated GDP is 0.5%. This means that the GDP deflator (“inflation”, if you must) is running at 9.5%.

    Prices change either because of supply or demand. So a price may go up because of a change in the supply and/or velocity of money (a demand-side factor) or because it’s become costlier to produce that good (a supply-side factor).

    In the LONG-RUN, all inflation is demand-side (a “monetary phenomenon) because the rate of output in industrialised countries is more or less steady over time; supply-shocks are temporary and self-adjusting insofar as the economy has market pricing. Price inflation is only not self-correcting when central banks deliberately increase NGDP/AD above the level of output, e.g. with inflation targeting.

    The kind of sustained stagflation that the US experienced in the 1970s was primarily the result of a demand-side factor i.e. money supply profligacy after the end of the Bretton Woods system. The stagnation from oil prices/falling productivity/etc. was separate and basically irrelevant as a cause of the Great Inflation, since average RGDP growth during the 1971-1981 period was around trend at 2.63%, while inflation and NGDP were averaging 7.3% and 10.12% respectively.

    If we had a Selgin-style Fed (which would be worse than free banking but better than now) which targeted NGDP at the long-term rate of output (2.5%-3%) then all inflation would be “cost-push” inflation i.e. supply-side driven and self-correcting.

  17. Gravatar of John John
    15. August 2011 at 15:15

    W. Peden,

    The point I was trying to make was that a supply shock, even to a commodity as important as oil, is insufficient to cause a general rise in prices. That rise has to come from the side of money. For another example, if the government suddenly added an extra 20% sales tax on every single item, all businesses wouldn’t suddenly be able to charge more for every item.

    Your NGDP example seems sort of silly. It a rise in oil prices caused production to drop, GDP would drop because of the drop in production (ie finished goods and services sold) not because of the price deflator. The purpose of trying to find real GDP is to measure increases in production by taking out inflation.

    If NGDP were growing constantly at 10% and consisted of 6% real growth and 4% inflation, then the oil shock would reduce real growth to 5.5% and if the central bank wanted to hit the 10% NGDP target they would have to pursue 4.5% inflation rather than 4%.

  18. Gravatar of W. Peden W. Peden
    15. August 2011 at 15:37


    It appears that the distinction was indeed verbal rather than practical, since as you lay out your position there I find it perfectly correct. I thought that you might be implying that the price of oil has some sort of direct relation with the velocity of money.

  19. Gravatar of George Selgin George Selgin
    16. August 2011 at 00:59

    My productivity norm would set nominal income growth at the rate of weighted (expected) labor and capital input growth, not that of output growth. This generates deflation at the TFP growth rate.

  20. Gravatar of W. Peden W. Peden
    16. August 2011 at 05:10

    George Selgin,

    Thanks for the clarification. That makes even more sense, since it means that prices are fully performing their Hayekian signalling message i.e. “telling” people that the economy is becoming more productive over time.

  21. Gravatar of Scott Sumner Scott Sumner
    16. August 2011 at 06:36

    Ben, I agree.

    John, You said;

    “You’re right that the Fed was responsible for consumer incomes during the boom years as their irresponsibly loose monetary policy (specifically 2003-4)”

    But monetary policy wasn’t particularly loose during 2003-04.

    This issue isn’t whether fast growth in incomes was good in Zimbabwe, of course everyone agrees it wasn’t. But to have a fruitful discussion we first have to agree on who drives income growth. I say the Fed, you argued something else. That was the point of my example.

    I agree with Keynes, as he would have made exactly the same argument for fast NGDP growth, which is equally bad.

    David, Yes, per capita NGDP growth targeting would be slightly better, but the difference between the two is very small from one year to the next, and we lack good population data (because of illegal immigration.

  22. Gravatar of John John
    16. August 2011 at 14:42


    I agree monetary policy (the Fed) drives nominal wage growth while increasing marginal productivity of labor due to capital investment drives real income growth. Real wage growth and nominal wage growth are often contradictory as productivity increases tend to drive down prices and rising prices encourage capital consumption.

  23. Gravatar of ssumner ssumner
    17. August 2011 at 14:36

    John, I agree.

Leave a Reply