NGDP Targeting v. Nominal Wage Targeting

This post follows up on an observation I made in my post about George Selgin’s recent criticism of John Taylor’s confused (inasmuch as the criticism was really of level versus rate targeting which is a completely different issue from whether to target nominal GDP or the price level) critique of NGDP targeting. I found Selgin’s discussion helpful to me in thinking through a question that came up in earlier discussions (like this) about the relative merits of targeting NGDP (or a growth path for NGDP) versus targeting nominal wages (or a growth path for nominal wages).

The two policies are similar inasmuch as wages are the largest component of nominal income, so if you stabilize the nominal wage, chances are that you will stabilize nominal income, and if you stabilize nominal income (or its growth path), chances are that you will stabilize the nominal wage (or its growth path). Aside from that, the advantage of NGDP targeting is that it avoids a perverse response to an adverse supply shock, which, by causing an increase in the price level and inflation, induces the monetary authority to tighten monetary policy, exacerbating the decline in real income and employment. However, a policy of stabilizing nominal income, unlike a policy of price level (or inflation) targeting, implies no tightening of monetary policy. A policy of stabilizing nominal GDP sensibly accepts that an adverse supply shock, by reducing total output, automatically causes output prices to increase, so that trying to counteract that automatic response to the supply shock subjects the economy to an unnecessary, and destabilizing, demand-side shock on top of the initial supply-side shock.

Here’s Selgin’s very useful formulation of the point:

[A]lthough it is true that unsound monetary policy tends to contribute to undesirable and unnecessary fluctuations in prices and output, it does not follow that the soundest conceivable policy is one that eliminates such fluctuations altogether. The goal of monetary policy ought, rather, to be that of avoiding unnatural fluctuations in output–that is, departures of output from its full-information level–while refraining from interfering with fluctuations that are “natural.”

The question I want to explore is which policy, NGDP targeting or nominal-wage targeting, does the better job of minimizing departures from what Selgin calls the “full-information” level of output. To simplify the discussion let’s compare a policy of constant NGDP with a policy of constant nominal wages. In this context, constant nominal wages means that the average level of wages is constant, any change in a particular nominal meaning an equivalent change in the relative wage. Let’s now suppose that our economy is subjected to an adverse supply shock, meaning that the supply of a non-labor input has been reduced or withheld. The reduction in the supply of the non-labor input increases its rate of remuneration and reduces the real wage of labor. If the share of labor in national income falls as a result of the supply shock (as it typically does after a supply shock), then the equilibrium nominal (as well as the real) wage must fall under a policy of constant nominal GDP. Under a policy of stabilizing nominal wages, it would be necessary to counteract the adverse supply shock with a monetary expansion to prevent nominal wages from falling.

Is it possible to assess which is the better policy? I think so. In most employment models, workers accept unemployment when they observe that wage offers are low relative to their expectations. If workers are accustomed to constant nominal wages, and then observe falling nominal wages, the probability rises that they will choose unemployment in the mistaken expectation that they will find a higher wages by engaging in search or by waiting. Thus, falling nominal wages induces inefficient (“involuntary”) unemployment, with workers accepting unemployment because their wage expectations are too optimistic.  Because of their overly-optimistic expectations, workers’ decisions to accept unemployment cause a further contraction in economic activity, inducing a further unexpected decline in nominal wages and a further increase in involuntary unemployment, producing a kind of Keynesian multiplier process whose supply-side analogue is Say’s Law.

So my conclusion is that even nominal GDP targeting does not provide enough monetary stimulus to offset the contractionary tendency of a supply shock.  Although my example was based on a comparison of constant nominal GDP with constant nominal wages, I think an analogous argument would lead to a similar conclusion in a comparison between nominal NGDP targeting at say 5 percent with nominal wage inflation of say 3 percent.  The quantitative difference between nominal GDP targeting and nominal wage targeting may be small, but, at least directionally, nominal wage targeting seems to be the superior policy.


39 Responses to “NGDP Targeting v. Nominal Wage Targeting”

  1. 1 W. Peden December 9, 2011 at 11:37 am

    Great post. I shall wait to see informed responses first, but I’m convinced so far.


  2. 2 Peter December 9, 2011 at 11:38 am

    I assume you would have a similar problem if a country discovered large quantities of oil which caused the wage share of NGDP to fall.


  3. 3 Peter December 9, 2011 at 11:40 am

    But do we know that wages would fall as share of NGDP if we did NGDP targeting? Could it be that it’s the inflation target that causes that today?


  4. 4 Bill Woolsey December 9, 2011 at 12:16 pm

    I think you are correct in general, but I don’t think you get the secondary multiplier. How is that consistent with nominal income remaining constant?

    Bad havest. The price of corn rises and the output of corn falls. The price level rises and output falls by arithmetic.

    The real wage falls (because labor purchases less corn and the same amount of everything else.)

    You assume that labor’s share falls. so that real wages must fall more than in proportion to real output. The price level rose in proportion to the fall in real output, and so real wages only fell in proportion to real output. That was too little.

    Prices need to rise more or else nominal wages must fall. Suppose that doesn’t happen and so employment falls. Workers are poorer and spend less. Firms sell less. So they produce less and lower prices. Nominal income falls below target.

    It seems to me that this secondary effect you describe is in error, because nominal income falls below target. Nominal income targeting keeps that from happening.

    By the way, since I don’t understand why the bad harvest reduces labor’s share exactly, why not just start there. (Were you thinking about oil? And that the people who own oil wells raise their share of income? While there is some truth to that, for the U.S. anyway, where oil is largely imported, I am not sure that matters much.)

    If nominal wages are especially sticky, then anything that requires a shift in real wages would result in less disruption if nominal wages are targeted, and output prices and other nominal incomes fluctuate.

    While profits are obviously very flexible, a good bit of capital income is pretty sticky too.

    Making prices and profits rise enough so that there is fully employment at the ttarget nominal wage doesn’t sound too bad, though I suppose all of those debt holders will complain. I think Rowe was pointing out that there might be political issues with forcing down profits enough to make sure that too many people didn’t get to large of a raise.

    Also, as I argued last time, there is a problem with using a wage index. If the monetary authority is targeting it, using some kind of feed back based on its actual or expected future level, it is going to get really sticky and it will provide poor signals of disequilibrium. This is especially true of expectations.

    You would do much better to go with some kind of measure of aggregate wage income. If that gets sticky, then there is no problem. You have both wages and employment covered. Say, total wage income grows 3% or 4% per year. Or even per capita wage income grows 1%.

    But if you do wage rates, and employment fluctuates, then you are depending on wages deviating from target to signal the problem. But they will tend to stick to the target. (I see it as an implication of Goodhart’s Law.)


  5. 5 Carl Lumma December 9, 2011 at 1:08 pm

    Since NGDI = NGDP, nominal wage targeting and NGDP targeting aren’t going to differ greatly; both can artificially restrict natural response to a supply shock. TIPS-spread beta targeting should do better. See here for more:


  6. 6 David Pearson December 9, 2011 at 8:20 pm

    Which nominal wage trend is appropriate? Did nominal wages decelerate in the 90’s sometime? Did they lag NGDI, especially in the past decade? If the trend changed, what would have been the implication of continuing to target the previous trend? Further, if we are going to exclude the 2007-2011 period from the trend calculation, then shouldn’t we also exclude all previous recessions, since they were similarly caused by a failure to target NGDP?

    What is the nominal wage trend in Japan? If Japan has another ten years of decent per-capita real gdp growth and low unemployment, would you still wish to “push” nominal wages back to their 1980-1990 trend? What about Italy? It has had slow nominal wage growth for a decade now. Is the optimal “trend” there the pre-Euro trend? If so, what unemployment rate resulted from that pre-Euro trend?

    Establishing a general rule for trend calculations seems quite problematic.


  7. 7 Bill Woolsey December 10, 2011 at 6:32 am

    In the US, the trend growth rate of nominal compensation per worker during the Great Moderation was 3.96%. The growth rate of nominal wages for production and nonsupervisory employees was 4.08%. The trend growth rate for aggregate labor compensation was 5.3%, which is only slightly greater than nominal GDP, which is 5.24%.

    Of course, Glasner’s point would be about fluctuations.


  8. 8 David Pearson December 10, 2011 at 10:01 am

    “Of course, Glasner’s point would be about fluctuations.”

    Every fluctuation is assumed to be caused by a temporary shock rather than an adjustment to a new trend line. If I have a hammer (trend targeting), every problem looks like a nail.


  9. 9 The Liquidationist December 10, 2011 at 10:33 am

    Very interesting post.

    However I think the comparison between NGDP and nominal wages targeting reveals the weakness at the core of all AD stabilizing regimes when looking at demand rather than supply shocks.

    In the face of a demand shock either type of targeting regime will successfully stabilize AD as long as interest rates are able to bring about equilibrium in the money markets. However if we hit the situation where AD (and in particular the demand for labor) are so depressed that even at the zero-bound the money market does not clear at a high enough level to maintain AD then we can immediately see a problem with nominal wage targeting.

    In this scenario employment will be at lower than desired levels. This will require moving the supply of labor curve to the right (and wages downwards), or (if one assumes sticky wages) a movement down the existing supply curve towards the new equilibrium . Any tendency in this direction would be explicitly prevented by nominal wager targeting (and implicitly so by NGDPT).

    So at the zero-bound any forms of targeting that aims to stabilize AD can work only if either the CB can magically shift the demand for labor curve to the right via expectations settings, or via inflation that will create negative real interest rates.


  10. 10 Bill Woolsey December 10, 2011 at 1:05 pm

    Nominal GDP targeting doesn’t prevent nominal wage rates from falling. To the degree nominal wages do fall, the expectation that nominal GDP will rise back to target sooner or later reinforces the increase in the quantity of labor demanded that exists given the current, perhaps depressed level of nominal GDP. Labor is relatively cheap.

    As for the zero nominal bound, market monetarists favor negative interest rates on reserve balances, and quantitative easing in whatever amount is necessary. If purchasing all the safe and short government bonds isn’t enough, you move to riskier and longer term to maturity bonds. Are all interest rates supposed to be zero?

    What is the alternative? Shifting down supply curves or having the supply curve shift right doesn’t do any good unless this somehow generates additional real expenditure to purchase the added output produced by the labor. I don’t know what you have in mind, but stable nominal GDP expectations beat the pigou effect.


  11. 11 Benjamin Cole December 10, 2011 at 1:52 pm

    My complaints lately about the Market Monetarism movement is that we are moving towards splitting hairs with each other, instead of concentrating on a platform, and language, that will capture the policy process.

    The language of “nominal wage targeting” alone is toxic. Sounds like a socialist pipe-dream, whatever the merits.

    I advocate Market Monetarists develop a concrete program—not perfect in everyone’s eyes, but good enough to hold your nose and support—instead of ever-more esoteric communiques.

    My rough-house suggestion is that we back a QE program of $100 billion a month until certain NGDP targets are met, such as 7 percent NGDP growth, and limit IOR. All this clearly communicated to the public by the Fed. We also say we generally support lower taxes and regulations on business and we are pro-business and want American to work and get off the dole.

    Why the concrete $100 billion number? It is platform and policy the public can understand. It will give assurance to the markets that action will be taken, not just plans and words.


  12. 12 The Liquidationist December 10, 2011 at 3:15 pm

    I agree that paying interest on reserve balances makes no sense in the current situation but fail to see how QE can be anything other than inflationary Is QE not a direct transfer of new cash to those whose assets are bought above free-market value ? How does this shift either the demand or supply curve for labor ? Perhaps it somehow shifts the demand for labor curve by expectations settings – but if so it does this via a mechanism I have never seen convincingly explained.

    In response to : “Shifting down supply curves or having the supply curve shift right doesn’t do any good unless this somehow generates additional real expenditure to purchase the added output produced by the labor”

    This would reduce production costs and increase profits and so increase the demand for labor. Surely Say’s law tells us there would be no problem finding purchases for the additional output?


  13. 13 Benjamin Cole December 10, 2011 at 3:53 pm

    The Liquidationist:

    I am not sure your excellent comment is directed at me, but I will answer it. Let me lead by saying I am not a Phd economist.

    My take on QE is that it turns a lot of tricks at once.

    1. It puts cash into the hands of people who before held often-inert assets. We can hope they invest in remaining assets (raising values, especially real estate) or spend it, raising NGDP.

    2. I hope it is somewhat inflationary. Sticky wages etc. Also helps deleverge taxpayers. If we run a balanced budget but five percent inflation for five years, we cut down our debt by almost one-third.

    3. No one is supposed to say this out loud, as it falls under the banner of “monetizing your debt.” The fact is, we deleverage taxpayers when we print money and QE and buy Treasury bonds, or in the private sector if we buy commercial IOUs. This should be inflationary, but the problem in Japan and perhaps the USA is not inflation, but deflation, or too-low inflation.

    The USA of the 2010s is far different for that of the 1970s. We have global supply chains for goods, labor, services and capital. Unions are dead. The inflationary trends are muted, even dead in such a scenario. When $800 billion in Benjamin Franklins is healed offshore, I no longer know what money supply means. Or when wealth is recorded as blips on computer chips. Your claim on the output or wealth of our society is just blips on a computer chip.

    The problem is, there is not enough blips out there to boost demand.

    What if drug-lords decided to repatriate the $800 billion in Ben Franklins held offshore and spend it? We would have a boom. No right-winger is ever against that—but printing up $800 billion and having a boom is wrong. It all begins to seem a bit koo-koo after a while.

    My morality is that prosperity is better than poverty, and inflation is much less important than prosperity.

    The historical record is that from 1982 to 2007, the USA prospered with mild inflation in the 2 percent to 6 percent range. During this time, our industrial production doubled. Doubled!!!

    Why obsess about 2 percent inflation rates today? So we have 5 percent, and the economy get hot.

    That’s fine with me.


  14. 14 Bill Woolsey December 10, 2011 at 4:06 pm

    An increase in the quantity of money increases money demand for output and so the derived money demand for resources, including labor. It raises the equilibrium money wage for labor.

    The effects of an increase in the quantity of money on both real and nominal interest rates is ambiguous. While the direct effect of the Fed’s purchase is to raise the price and lower the yield, the expectations of higher demand for output will tend to raise both real and nominal interest rates. How this works is other asset holders sell more than the Fed buys and instead purchase consumer or capital goods.

    Of course, asset prices could rise and yields fall. But there are the free market prices and yields. The lower prices and higher yields only exist because the central bank is failing to expand the quantity of money, which it monopolizes, to meet the demand to hold it. The shortage of money creates a liquidity effect that keeps yields from falling.

    As for Say’s law, you need to do better than that. Lower wages results in more profit, but selling the extra output lowers prices, so that profits return to the initial level. If you assume that the nominal quantity of money is fixed, then the lower prices raise real money balances. As those surpass the amount people want to hold, some of them are lent, lowering interest rates. If the nominal interest rate is already zero (your assumption,) then what is left is an increase in real wealth, less saving, and more consumption. That is called the “pigou” effect. Unfortunately, that only works with outside money, like gold. For money that is an asset to holders and a debt to issuers (including taxpayers,) the lower price level transfers wealth.

    With nominal GDP targeting, the lower wages allow for lower prices, which allows for more real sales given nominal spending on output. But, of course, the quantity of money is also rising, raising nominal spending as well. And so, lower prices and wages and higher nominal expenditures both raise real output and employment.


  15. 15 The Liquidationist December 10, 2011 at 6:08 pm


    first thanks for being prepared to engage in what I find a very interesting discussion.

    On the derived demand for labor. Other things being equal wouldn’t an increase in the money supply be neutral and just increase prices of all goods with no particular effect on demand for labor ? I can see that when the increase comes as a result of additional bank lending (which is likely to go into investment) then this will increase demand for labor, but when it comes from QE (which seems aimed at forcing any kind of spending) then I see no reason to believe it will increase investment relative to consumer spending and so do not see any net benefit on demand for labor.

    On Says law: I hope I am not missing anything but I really think it is that simple, No matter what the current IR and without the need for any Pigou effect then a fall in wages should not cause a problem. Firstly (depending upon the shape of the curves) a fall in wage-rate may actually increase the total money spent on wages and increase AD. However, even if one assumes that the total wage bill falls as a result then revenue from wages + profits should still equal sales revenue so all markets should be able to clear.

    It strikes me that perhaps the theory behind MM is somehow assuming that depressed AD at the zero-bound causes prices not just to move to a short-term equilibrium we don’t like because it yields a low level of resource usage but that actually it yields a non-equilibrium position where demand for labor and other goods is being thwarted by sticky prices – hence the need for an increase in the money supply via processes such as QE to enable all supply to again be bought up at current price levels. .

    If that is the case then I disagree with it. Investment is low because of perceived lack of investment opportunity and unemployment is high because of the (government created) shape of the supply curve. At the zero-bound the choice is between years of recession, or addressing the structural issues that lie at the heart of this crisis.


  16. 16 The Liquidationist December 10, 2011 at 6:16 pm


    I was actually replying to Bill, but your reply was interesting. It confirmed my view that perhaps Market Monetarism is gaining popularity amongst those for whom inflation is seen as part of the solution.


  17. 17 David Glasner December 10, 2011 at 7:41 pm

    W. Peden, Thanks. I believe that Bill Woolsey is correct that my comment about a secondary multiplier effect under NGDP targeting when nominal wages are falling is nearly correct. There might be a small secondary multiplier effect, but stabilizing NGDP would prevent the secondary multiplier from have a large cumulative effect.

    Frank, It’s not my definition or Alan Greenspan’s it’s the definition in every econ textbook. Paying people not to work reduces the amount of hours worked and total output so the ratio may go up or down as a result. What Humphrey Hawkins did was to make it clear that the Fed has a responsibility to keep employment high as well as prices stable. Keeping employment high is not the same as maximizing productivity which is the ratio of input to output. Employment is the amount of labor input and GDP is total output. As employment increases it’s not clear whether the ratio of output to labor input is going up or down, but beyond some point, the ratio almost certainly starts to fall (which is not necessarily bad).

    Peter, I don’t know why you would think that. On the contrary an increase in the amount of oil available would probably increase the real wage and labor’s share. My point was not that every supply shock reduces labor’s share, but that it is likely that some supply shocks have that effect as has been true in the past when the price of oil has gone up rapidly.

    Bill, See my comment above to W. Peden, so I think that we are basically in agreement. See also my comment to Peter. I was not thinking of a bad harvest, but an increase in oil prices. Even though we import a lot of oil, we still produce a lot domestically and other the prices of other domestically produced fossil fuels are highly correlated with oil prices. I don’t follow your point about sticky wages, expectations and nominal wage targeting, so you will have to help me out some more before I can respond.

    Carl, Thanks for the reference. I have a lot of reading to catch up on.

    David, Why should I care about changes in the trend of nominal wage growth. The point of stabilizing nominal wages is so that labor markets remain close to equilibrium and that adjustments are shifted to markets for output and other factors of production. Slower real wage growth means faster inflation (or slower deflation) faster real wage growth means slower inflation (or faster deflation). The problem in the eurozone is that the exchange rate has been eliminated as a mechanism for achieving adjustments in real wages across countries. Slower growing countries have to have real wages that fall over time relative to wages in faster growing countries. If nominal wages are sticky, then the only way for the adjustment to take place in the eurozone is for the rate of inflation in the eurozone to be fast enough so that nominal wages in the slowest growing countries don’t have to fall. Is Mrs. Merkel ready to sign on to that one? If not, the euro cannot survive.

    Bill II, Where did you get those numbers from?

    David, Why does a change in the trend matter if nominal wages are being stabilized. The adjustment takes place in terms of the price level, which is less disruptive than a change in the wage level.


  18. 18 David Glasner December 10, 2011 at 8:15 pm

    Liquidationist, You seem to have in mind a sudden wave of pessimism that reduces private investment, causing a drop in aggregate demand. If equilibrium requires a negative real interest rate, a stable nominal wage level will produce an increase in the price level and an expected rate of inflation that reduces the real interest rate sufficiently to maintain equilibrium and prevent involuntary unemployment. So I don’t understand why you think that nominal wage targeting (or nominal GDP targeting) wouldn’t work in this case.

    Benjamin, You are just way too practical. Why are you always trying to force ivory tower types like me to live in the real world?

    I’ll have some further responses tomorrow.


  19. 19 The Liquidationist December 10, 2011 at 10:56 pm

    I agree that in theory wage targeting in this scenario will lead to inflation and that this inflation will create a negative real rate of interest which will cause people to invest rather than hold cash and result in the level of employment increasing. I have never seen that acknowledge before as part of the MM framework.

    My objection is that inflation causes problems with economic calculation that will lead to distortions in the structure of production and that it is better to address now the issues with investment (often referred to as ‘regime uncertainty’) and the labor market (UI and min wage) as it will be tougher to do so after inflation has been injected back into the system.


  20. 20 David Glasner December 11, 2011 at 1:01 pm

    Liquidationist, The point that I have been trying to get across, and I think that many if not most Market Monetarists hold a similar view, is that the optimal rate of inflation is not a constant over time. Therefore targeting a particular rate of inflation — zero, two-percent, or whatever — will eventually get you into trouble. In this post, I suggested that you are less likely to get into trouble by targeting nominal wages or the rate of growth of nominal wages. I can’t remember if you have identified yourself as a follower of Austrian economics, but I suspect that you are at least partially under their influence. So I would point out to you that Hayek recognized in the early 1930s that stabilizing nominal income would, in theory, give you the optimal monetary policy (by making money neutral, i.e., allow real income to be determined by the underlying real forces of the economy). But stable nominal income implies that prices rise when real output is contracting and fall when real output is expanding. So there is a close connection between the policy I am suggesting (and the similar policy of NGDP targeting advocated by most Market Monetarists) and Hayek’s conceptual benchmark of stable nominal aggregate expenditure.

    Bill and Benjamin, I generally agree with what you are saying, though, as you know, that doesn’t mean that we agree on every detail.


  21. 21 The Liquidationist December 11, 2011 at 3:37 pm

    I am very much influenced by the Austrian school but as a supporter of Free Banking I understand the benefits of a regime that delivers stable AD.

    I understand the distinction you are making in your (excellent) post and see that in normal times nominal wage or nominal income targeting will be better than inflation targeting in a CB model.

    In a free banking model the way that AD is stabilized is via banks increasing lending in response to an increase in the demand for money (which has the effects of increasing banks reserves) This is somewhat similar to the way a CB might expand the money supply when they need to “loosen” monetary policy. My concern is this (It applies to a free-banking world as much a CB-centric one): When the economy is in a deep recession like the current one then business will not want to borrow money even at very low rates of interest and it will be impossible to stabilize AD via increased lending (“conventional means” in a CB model)..

    In this situation I think that given the choice between increasing the money supply via QE or other unconventional methods that I believe can work only via using inflation to give a negative real interest rate , or taking the tough option and forcing through the supply-side changes that will allow the economy to lift itself out of this morass then the time has come to consider the latter option.


  22. 22 Bill Woolsey December 12, 2011 at 6:32 am


    While banks can expand credit by lowering the interest rate they charge and then respond to the demand for loans, they can also expand credit by purchasing bonds. If banks lower the interest rates they charge on commercial loans and find few takers, they can still purchase bonds. The quantity of money rises.

    With a central bank, the parallel process would be discount poicy. The central bank can lower the interest rate it charges for loans to commerical bank and then respond to the demand for loans by banks. However, a central bank can also understake open market purchases–purchase bonds. This is quantitative easing.

    The Fed uses open market operations all the time. But they don’t pay any mind to what this does to the “quantity” of base money, and simply look at the Federal funds rate. “Normal” monetary policy is about manipulating short term interest rates. After the rate fell below .25 percent, they began to talk about what sort of bonds they were buying and not talking about what they think that will do to the Federal Funds rate.

    Also, the efforts to manipulate the Federal Funds rate has traditionally involved trades in Treasure bills, if not repurchase aggreements–overnight lending or borrowing by the Fed using the security of assorted government bonds. If the interest rate on T-bills is zero, (or very near zero,) then I think it is likely that open market purchases will have little effect. While the quantity of money will rise, the demand to hold money will rise in proportion. And so, if one thinks of conventinal policy as the Fed increasing the quantity of money as purchases of T-bills, then that won’t be effective. However, the Fed’s balance sheet has long included other sorts of government bonds than T-bills. And so, having the Fed buy long term government bonds is nothing particularly new. But advertising this as being monetary policy is new. Quantitative easing. The Fed is also buying goverment guaranteed mortgage backed securities in very large amounts, and that is new.

    Free banking just cuts out the central bank, and open market purchases are still a possible means of expanding the quantity of money. I am not sure why you focus so much on commercial loans.

    As for “supply side” policies–these are unlikely to work well if the reason for the depressed expecations is poor expected sales. The primary effect of improved supply side policies is making firms even more frustrated by the problems in making sales.

    How this works out in the long run is lower prices and wages, higher real balances, and more real expenditure. If prices are expected to recover, then lower current prices and wages generates expecations of future inflation and negative real interest rates. If the lower price level is permanent, then this doesn’t happen. The lower price level and lower wages cause higher real balances and more real wealth, which reduces saving, and raises real consumption. The lower saving is an increase in the natural interest rate.

    Anyway, improved supply side policies result in this equilibirum involving even lower prices and wages and higher output.

    Expanding the quantity of money in this situation doesn’t necessarily require inflation. It raises the equilibirum price level, but if the current price level is above equilibirum, then it just means less deflation is necessary.

    On the other hand, if some prices are flexible and others sticky (including wages,) then an expansion in the quantity of money raises the flexible prices back up, and avoids the need for the sticky ones to fall. And so, there is inflation–a reverse of the deflation.

    If there has been a productivity shock, then nominal GDP targeting does raise the price level.


  23. 23 The Liquidationist December 12, 2011 at 11:01 am

    That is a very good point about bonds. Indeed free banks may well increase purchases of these if the demand for (or return from) commercial loans is low. Banks would always maintain a balance between loans and bonds that would keep equal the risk-adjusted returns on these assets. This would clearly allow free banks to expand the money supply further but I am not convinced it would be equivalent to QE as banks would stop buying bonds when the expected returns became too low, whereas QE (as suggested by Market Monetarists) would keep on buying bonds and other assets (no matter what the risk) until NGDP returns to trend.

    On the supply side: I believe that any changes that increase the supply of labor at a give wage rate (moves the supply curve to the right) will surely move the demand for labor along the demand curve and increase the equilibrium level of employment (subject to normal elasticity on these curves) I am totally missing why there would be any problem selling the higher output resulting from this even if AD is below its trend level.

    On QE: If I understand correctly the process is supposed to work as follows

    – Carry out QE until AD moves sufficiently to the right.
    – This increase in AD will initially affect only demand for consumer goods (since if business needed funds to invest they could borrow at low rates anyway)
    – The increase in demand for consumer goods will cause expectations of future inflation
    – The expectations of future inflation will reduce real interest rates below zero which will move demand for investment funds along the curve and increase investment
    – Somehow there will be no actual inflation because this will be offset by the deflation that would otherwise have occurred.

    Is that a correct understanding ?

    If so then (apart from the last part on the deflation offsetting the inflation that I don’t really get) I think this process may indeed cause investment to increase short-term but it seems that the only mechanism is inflation expectations causing negative real interest rates and this still sounds like a very dangerous strategy to me.


  24. 24 David Glasner December 13, 2011 at 9:27 am

    Liquidationist, If the natural rate of interest is negative, why shouldn’t you have inflation to adjust the market rate to the natural rate?


  25. 25 The Liquidationist December 13, 2011 at 3:15 pm

    That’s a good question. In my view this has the following dis-advantages

    1. General dis-coordination effects of inflation for economic calculation
    2. While inflation does create negative real interest rates it also decreases the incentive to save. This adds to the overall bias that QE shows to increased spending on consumer goods rather than capital goods.

    These effects combined means that the “equilibrium’ created by QE-induced inflation may be short-term and inferior to an equilibrium where the supply curve for labor moves to the right in response to the zero-bound situation.


  26. 26 David Glasner December 13, 2011 at 6:25 pm

    Liquidationist, It is not clear to me why inflation necessarily involves dis-coordination effects for economic calculation. Even if it did, it is not clear that those effects are worse than the loss of output associated with a market interest rate above the natural rate. When the natural interest rate is negative saving is socially wasteful at the margin, so why do you want to encourage it?


  27. 27 The Liquidationist December 14, 2011 at 8:07 am

    I am comparing 2 different equilibriums

    – The non-interventionist one we get if the supply curve moves to the right when interest rates approach zero
    – The one we get from QE when the AD curve for consumer goods moves to the right, induces inflation and leads to a equilibrium at negative IRs

    Other things being equal the first (as it reflects real consumer preference) looks to me that it is clearly preferable. The question then becomes what are the costs of addressing the supply side issues that appear to make option 1 slow to occur v the costs of inflation and market distortion caused by option 2.


  28. 28 David Glasner December 14, 2011 at 1:13 pm

    Liquidationist, Explain to me what the market mechanism is that achieves an equilibrium when the market rate is above a negative natural rate. You can’t just wave your hands and cause the AS curve to shift to the right.


  29. 29 The Liquidationist December 14, 2011 at 9:11 pm

    – There is a decline in the demand for money which causes AD ( and therefore the demand for labor curve) to shift left.
    – Initially the supply of labor moves along the supply curve to meet the shifted demand curve. Wage rates fall.
    – Other goods will also move along their supply curves in response to the new demand for money which will decreasing their prices
    – As a result of these falls in other prices suppliers of wages will observe that a given wage will now buy more and cause the supply curve to shift right
    – Supply curves for other goods will likewise move and we will have an iterative process of shifting supply curves until we hit a new monetary equilibrium (which may be defined as all supply curves reflecting an accurate view of the purchasing power of money)


  30. 30 Bill Woolsey December 15, 2011 at 4:39 am


    I presume you mean the demand for money _increases_, which reduces aggregate demand.

    Why does the AD curve have a negative slope? (Why does a lower price level for final output lead to an increase in real expenditure on output?)

    Anyway, so with the lower demand for output, there is also a lower demand for labor. And so, wages and the quantity of labor decrease. The lower wages imply a decrease in costs, so firms cut prices.

    The lower price level raises the quantity of goods demanded. So, you see, that we are appealing to the negative slope of the AD curve. Why does a lower price level raise real expenditure on output?

    Anyway, the short run aggregate supply curve shifts to the right because workers are willing to accept lower wages. I suppose this could be because they recognize that lower prices mean that an given nominal wage is a higher real wage. The lower wage lower costs further, and so prices fall even more. The shift in the short run aggregate supply curve to the right moves along the aggregate demand curve. The lower prices are resulting in higher real expenditure. Firms produce more and hire more workers.

    This is how the lower wages results in a higher quantity of labor demanded.

    So, all of this argument requires that the aggregate demand curve has a negative slope with regard to the price level. Lower price level, higher real expenditure.

    With nominal GDP targeting this occurs by definition. A lower price level, given the flow of money expenditures on output, results in a higher level of real expenditure.

    Without nominal GDP targeting, there has to be some other argument.

    The orthodox argument is that given the nominal quantity of money, a lower price level raises real money balances. As real money balances rise beyond the amount people want to hold, the excess balances are spent on output.

    This is called the real balance effect. How the real balance effect works depends on the nature of the money. if it is inside money, like checking accounts issued by banks, then the lower prices do make those holding money better off, but it makes those who issued the money, the banks directly, and those with bank loans indirectly, worse off.

    However, even though people are no better off on the whole, the increase in real balances suggests that people are holding too much money relative to other assets. As they spend on those other assets, the asset prices rise and interest rates fall.

    To the degree that money is of the “outside” form, which would include gold reserves with a gold standard, then the lower price level makes those holding the money better off and no one worse off. They are wealthier and so are motivated to save less. And so they consume more. This is called the Pigou effect.

    If nominal interest rates are already zero, then the ones that are zero can’t go any lower due to the real balance effect. That appears to leave only the Pigou effect. But really, that depends on there being one nominal interest rate that is at zero. In reality, there are lots (and all sorts of real assets) whose yields aren’t zero. So, the real balance effect can drive those yields down as well.

    Also, to the degree the increase in money demand starting all of this was temporary, then the lower level of prices and wages is also temporary. The expectation that prices will rise again is expectations of inflation. This reduces the real interest rate associated with any nominal interest rate, including the zero ones.

    I don’t see how this is any better than an increase in the nominal quantity of money. if “the problem” is that some nominal interest rates are zero, then the expansion in the quantity of money should involve purchasing financial assets that don’t have zero interest rates. As we have seen, the real balance effect is going to involve lower nominal interest rates.

    All that is given up is the Pigou effect. To the degree there is some outside money (government fiat currency in our world, though that is questionable,) an expansion in the nominal quantity of money depends on lower nominal interest rates more and doesn’t have the pigou effect resulting in more consumption. Of course, lower nominal interest rates also motivate more consumption too.

    If I had to critique your approach, the first would be possible confusion about aggregate demand. The quantity demanded for a single good is negatively related to price because the good gets cheaper relative to other goods, and so people buy it rather than other things. This substitution effect can’t raise the demand for everything.

    There is also an income effect. For normal goods, a lower price raises real income a bit, and so results in more demand. But this assumes constant money income (what nominal GDP targeting provides.) In reality, lower prices raise the real income of buyers and lowers real income of sellers. This is why we are left with the real balance effect–the given nominal quantity of money is worth more when prices fall. If the flow of money income was the same, then lower prices raise real income too. But if that is not true, the lower prices lower the dollar value of income. Sellers are earning less.

    No, aggregate demand is not like the demand for a single good, it is the other side of the coin of the quantity of money and the demand to hold money. The lower wages is not about somehow clearing the labor market. The entire disequilibrium in the labor market and output markets is due to a failure of prices and wages to adjust enough for the real quantity of money to match the increased demand for money. And careful consideration of how the increase in the real quantity of money causes real expenditure to rise again shows that it is very similar to the impact of an increase in the nominal quantity of money–just slower and more painful.


  31. 31 David Glasner December 15, 2011 at 5:49 pm

    Liquidationist, I generally agree with what Bill has to say (I am hedging becuase I did not read it as carefully as I ought to have and because I noticed one or two relatively minor points on which I might have a different view). I will just emphasize a point that I think Bill would agree with, but which I don’t think he focused on sufficiently, which is that you can’t talk about a supply curve for labor (or any other durable good) without at least implicitly making an assumption about the future wages and prices that are expected. How much labor is supplied today at a given wage rate depends on the expectations of workers on future wage rates. The higher (lower) expected future wage rates are relative to current wage rates, the less (more) labor will be supplied at a given wage rate in the present. Because workers are usually slow to revise their expectations of future wage rates, a deflationary shock that reduces current wage rates tends to raise expected future wage rates relative to current wage rates, causing a downward shift in labor supply in the current period. This is the microeconomics behind Keynesian involuntary unemployment. Because of incorrect expectations of future wages causes a downward shift in labor supply, Say’s Law implies that aggregate demand also falls as a result of the fall in supply. That’s why a policy of stabilizing the nominal wage, so that workers expectations of future wages are (on average) never incorrect, minimizes involuntary ( inefficient) unemployment.


  32. 32 The Liquidationist December 15, 2011 at 9:25 pm

    Thanks for your detailed analysis.

    Yes, that was a typo I meant increased demand for money.
    Regarding negative slope for AD curve: It is obvious that if prices fall and NGDPT stays constant then real expenditure must have increased. Similarly if prices fall and demand to hold money decreases as a result then clearly real expenditure must have increased.

    However it seems we are talking about the recovery following the consequences of an increase in the demand for money (our starting point is after the initial AD curve has shifted left).

    To achieve stability we need both monetary equilibrium and a price equilibrium that reflects underlying consumer preference. I think there are scenarios where NGDPT will fail to achieve this.

    As long as NGDPT works by the CB expanding the money supply via the banking system (and adjusting interest rates to align lending and borrowing) then it is simulating what the free market would do and more-or-less things will work well.

    When interest rates approach zero (which will happen if very high levels of perceived risk by businesses moves the demand for loans sufficiently far to the left) then I see the following differences between free banking and NGDPT:

    – Free banks would stop lending (and buying bonds) and accumulate reserves . This fall in the money supply would cause a monetary disequilibrium. This monetary disequilibrium would be solved by lower prices and also by relative prices adjusting to give bigger profit spreads. These increased profit spreads will in turn the loan market to cleari (and the money supply re-expanding). We will have both monetary equilibrium and prices that have adjusted to reflect underlying consumer preference in the balance between various forms of investment and consumption.

    – NGDPT would (based on previous comment) do the following:
    o Target assets that have non-zero interest rates in order to lower them. This would be distortionary because it disrupts the relationship between different interest rates that presumably represent different levels of perceived risk
    o Keep buying assets until the desired level of NGDP is achieved. This would be distortionary because the new spending would go primarily into consumer goods (since business could borrow at low IRs anyway if they wanted to invest more). This will lead to inflation and negative real rates (which has already been discussed in this thread) and a greater % of investment going into consumer goods than justified by underlying consumer preference .
    o These issues will cause a problem when QE ends and lead to the new NGDPT-derived equilibrium being short lived.


  33. 33 The Liquidationist December 15, 2011 at 9:53 pm


    I believe that assumptions about future expectations (about both wages and other factors such as future inflation and NGDP) are an important underpinning for Market Monetarism.

    I may need to think it through a bit more but it still seems to me that there is a contradiction between ensuring that workers expectations are never wrong by having a stable nominal wage, and the need for falling real wages that may arise due to increased risk-aversion by capitalists. I’m guessing that there is an assumption in there that risk-aversion will not vary as long as the CB targets the right things – but this still feels like a dangerous assumption to to me as long as the CB is not all-knowing.


  34. 34 Bill Woolsey December 16, 2011 at 4:08 am


    Having businessmen hold more money until wages fall relative to prices, and so “capitalists” earn enough more to compensate for risk doesn’t work. When they hire the workers back, wages rise back relative to prices.

    What they actually have to do is consume rather than save, and as capital goods wear out and are not replaced, the marginal product of capital rises and equilibrium real wages fall.

    One of the problems you describe with nominal GDP targeting (or nominal wage targeting, I guess) is that there is too much consumption. Whether or not consumption rises in that case depends on the interest elasticity of saving relative to the interest elasticity of investment. However, the Pigou effect, which occurs with a lower price level and outside money does involve an increase in consumption. Higher real balances, more real wealth, less saving and more consumption. It is that process, that allows capitalist/entrepreneurs to earn a higher return on investment to compensate for risk.

    With inside money, this doesn’t happen. Accumulating money just leads to lower interest rates one way or another. And lower interest rates results in more consumption and more investment.

    If you imagine that capitalist/entrepreneurs won’t accept lower interest rates, you are assuming the supply of saving is perfectly elastic with respect to the interest rate. And that means that they consume rather than accept a lower return. And, you can assume that this reduced willingness to take risk involves a decrease in the supply of saving. (It shifts up or to the left.) It is like the rich folks quit investing in production process and instead buy yachts and mansions. Accumulating money won’t do it. Oh… they might work less too. The supply of entrepreneurial labor drops as they retire (and enjoy the yachts and mansions.) The “wage” of the remaining ones is higher, and the wages of other types of workers fall. But to the degree it is their money we are talking about, accumulating money rather than investing, leaving aside the Pigou effect, just results in lower interest rates in the long run.

    I agree that quantitative easing shrinks the usual gap in yields between risky and safe assets. What is happening is the the monetary authority is taking on extra risk. It is issuing low risk base money while purchasing longer and riskier bonds. To the degree this makes the monetary authority more risky, this reduces the demand for base money and tends to relieve the monetary disequilibrium. If the situation persists, then the necessary quantity of money stays high. Quantitative easing persists. If, on the other hand, investors are willing to take greater risk to earn higher yields, then the demand for those riskier securities that the monetary authority holds is rising when the monetary authority needs to be selling them to reduce the quantity of base money, less of which is demanded.

    In my view, the first best solution is negative nominal interest rates on base money. If that is out, then having the monetary authority raise the quantity of base money and bear more risk is second best. A temporary deflation in prices and wages is much worse than either of those scenarios. Admittedly, I don’t think generating a higher trend inflation rate is much better than the deflationary option.

    P.S. Glasner is correct about future expectations being important. The most plausible source of pessimism is lower future nominal expenditure. But there are other possibilities.


  35. 35 David Glasner December 16, 2011 at 9:13 am

    Liquidationist, If the new equilibrium (or the adjustment necessary to achieving a new equilibrium) requires a reduction in real wages, under nominal wage targeting the reduced real wage is achieved by an increae in output prices. You seem to have an a priori objection to increased output prices under all circumstances, I don’t understand the basis for that objection. In principle, nominal wage targeting could be achieved by a fully automatic system of indirect convertibility, which is outlined in my book Free Banking and Monetary Reform.


  36. 36 The Liquidationist December 16, 2011 at 3:02 pm

    Firstly its probably worth stating that the reason I am so interested in changes in the rates of required returns by businesses is that without this arising in a scenario of decreased AD I do not see how one can get to a zero-bound. There would always be a positive rate of interest that clears the loan market and maintains NGDP without the need for QE.

    In regards to David’s point about increased output prices. I believe that increased profit spreads could indeed be achieved by a combination of falling wages and increased output prices. It is just forcing output prices upwards via QE that I am questioning

    I think Bill raises some great point that I have indeed not fully taken into account in my model. I am working on a more comprehensive version now which (if you guys are still interested) I will share when it is complete..


  37. 37 David Glasner December 18, 2011 at 7:02 am

    Liquidationist, Why are falling nominal wages less forced than rising nominal output prices? I look forward to seeing the next iteration of your model.


  1. 1 Browsing Catharsis – 12.10.11 « Increasing Marginal Utility Trackback on December 10, 2011 at 5:07 am
  2. 2 The three frontrunners for Fed chairman in a Romney administration Trackback on August 28, 2012 at 8:10 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: