The Chinese yuan: It’s not a zero-sum game.

I’d like to welcome any new readers from China, who may have discovered this blog through the Netease version.  I generally focus on U.S. monetary policy, but today I will discuss the Chinese yuan.  I should also mention that I will have to slow down for a few weeks, in order to revise a manuscript on the Great Depression that I am trying to get published.  If you are a new reader, there are plenty of older posts that discuss my view of how the current crisis has been misinterpreted.  Those views are quickly sketched out in the very first post, “About the blog.”  I will take a fairly long trip to China later this year—which might also lead to posts on topics related to China.

I’d like to consider the Chinese yuan from three perspectives:

1.  Long run trends in the real exchange rate

2.  The exchange rate and trade imbalances

3.  The exchange rate and world aggregate demand

1.  In my recent post on Estonia I discussed why very fast-growing economies tend to see their real exchange rate appreciate, due to what is called the “Balassa-Samuelson effect.”  This can occur either through a higher market (or nominal) exchange rate, or through a higher rate of inflation.  The intuition is simple.  Low income countries tend to have a lower cost of living, due to relatively low productivity in traded goods.  As they get richer their cost of living approaches developed country levels.

In my earlier trips to China I estimated the cost of living at barely a third of U.S. levels.  I expect China to eventually become a developed economy, at which time its cost of living should be similar to that of the US.  As a result, when the yuan was 8.28 to the dollar, I predicted the yuan would rise to about 3 to the dollar in the long run (perhaps 30 years.)  This surprised some of my Chinese friends, but it has already risen to about 6.8 in just a few years, and still has a long way to go.  I assumed the Chinese government preferred to make this transition through a higher nominal exchange rate, rather than high inflation.  In 2005 they seem to have made that choice.

In recent years, the US and European governments have pressured China to revalue the yuan upward.  Despite my views on the long run trend in the yuan, I think this is very bad advice, for both sides.  Indeed it might be better if the yuan was slightly devalued in the short run.  To see why, we have to consider two other issues that are widely misunderstood.

2.  The US currently runs a large trade deficit with China.  Many politicians in America seem to believe that deficit is caused by an undervalued yuan, and that the deficit hurts the US economy.  I think both views are wrong.  Like many other economists, I believe the trade deficit, or more precisely the current account deficit, simply reflects differential savings rates between the US and East Asia.  Countries that save more than they invest (China) will run current account surpluses with countries that save less than they invest (i.e. the US.)  The exchange rate does not cause imbalances, rather the real exchange rate moves to the level required to facilitate the necessary capital flows.

3.  I would go even further than many other western economists and argue that the US and Europe might have actually benefited from a yuan devaluation late last year.  To see why it will be necessary to move beyond the “zero-sum game” view of trade, the view that one country’s gains are another country’s loss.  When there is a steep worldwide drop in aggregate demand (AD), or nominal GDP growth, all countries suffer.  Any policy that is capable of boosting AD within China will also impact world AD, and thus has the potential of helping all countries.  First I’ll consider two theoretical ways of thinking about this issue, and then I will provide an example of where the US did exactly what they are telling China not to do.

If China devalued the yuan, and that boosted China’s growth rate, it would also boost growth in countries that supply inputs to China.  These include the more developed East Asian economies that supply manufactured inputs, capital goods exporters like the US and Germany, as well as commodity producers like Australia and Canada.

A more subtle, but potentially more important channel is through monetary policy.  Right now US monetary policy is limited by the fact that nominal rates cannot be cut below zero.  Although I have argued that there are other monetary policy channels, it would still help immensely if we could escape from the liquidity trap.  Faster worldwide economic growth, even if outside the US, will tend to raise the Wicksellian equilibrium interest rate in world capital markets.  If the equilibrium rate rose above zero percent, the current US fed funds target would become much more expansionary.  If it seems strange that higher rates could be expansionary, recall that the upswing in the US stock market since March has coincided with rising interest rates (real and nominal) in the 5 and 10 year T-bond market.  And this upswing also occurred about at the same time that many forecasters were becoming more optimistic about China’s growth rate for the rest of this year.

In 1933 the US tried the same sort of policy that I am now discussing in regards to China.  We were deep in depression at that time, and decided to sharply devalue the dollar against gold (and implicitly against most other currencies.)  The effect was immediate, and very positive.  Aggregate demand rose sharply on higher inflation expectations, pushing up industrial production by an amazing 57% between March and July 1933.  What many economists do not know, however, is that the dollar devaluation actually worsened our trade balance.  Exports rose about 3%, but imports soared 20%.  How did this happen?  Exports were held down by the weak economies outside the US, whereas the rapid growth in American industrial production raised US imports of key inputs, exactly the transmission mechanism that I hypothesized for a devaluation of the yuan.  In a deep slump, the income effect is often more powerful than the terms of trade effect.

There are many ironies in this story.  Although Europeans tended to oppose the sharp US devaluation in 1933, their economies actually benefited from the robust recovery that followed the devaluation.  Indeed if President Roosevelt had not made the mistake of introducing a high wage policy in July 1933, the Depression might have ended fairly quickly.  Today many Western leaders seem to favor a higher yuan.  But the last thing the world economy needs right now if for its most important growth engine to sputter because of a deflationary exchange rate policy.

Of course there are many political and economic factors that go into any exchange rate decision.  I don’t know whether depreciating the yuan would be a good idea right now, all things considered.  But I do think that western leaders make a mistake in pushing for a strong yuan, indeed I believe this is the same mistake they made with Japan in the 1990s.  But today with the entire world in a deep recession, the policy could hurt more than just one country.

Fortunately, because China continues to experience rapid productivity growth, even the recent policy of stabilizing the yuan is increasingly stimulative over time.  The Balassa-Samuelson effect is still in play and is gradually making China more competitive as its nominal exchange rate is currently fixed.  This may well cause China to be the first major economy to recover from the worldwide recession.  At some point in the future the yuan should and will resume its upward climb.  But not yet.  My purpose here is not to suggest exactly what should be done with the yuan, but to suggest a perspective that I think many in the West overlook.

PS.  Although I tend to agree with the Chinese government on its yuan policy, this does not mean I always agree with their economic policies.  I find Professor Yasheng Huang’s views on the need for faster economic reform in the countryside to be very persuasive.  I hope to discuss other aspects of the Chinese economy when I have had a chance to study the issues more fully.

PPS.  For those interested, here is a Netease page that includes both my blog and Greg Mankiw’s blog.  (Also here.) I am told that I am described as a “famous professor.”  That’s too polite (bie ke qi?)  Perhaps I will end up being better known in China than the US.



31 Responses to “The Chinese yuan: It’s not a zero-sum game.”

  1. Gravatar of Devin Finbarr Devin Finbarr
    23. May 2009 at 10:31

    In recent years, the US and European governments have pressured China to revalue the yuan upward. Despite my views on the long run trend in the yuan, I think this is very bad advice, for both sides.

    Right now, the main export of the United States is becoming its own currency. This is essentially the same position as Spain in the 16th century. Things are splendid in the short run. But how did it work out for Spain in the long run?

  2. Gravatar of ssumner ssumner
    23. May 2009 at 10:41

    Devin, I favor policies to sharply raise the US saving rate. That would address the problem you mention. Fiddling around with the nominal value of the yuan would not change its real value, and certainly wouldn’t change the fact that US saving is currently far below US investment.

    I think Spain’s main problem had nothing to do with gold outflows, but rather bad “supply-side” policies. Their public policies were less pro-growth than those in Britain.

  3. Gravatar of Devin Finbarr Devin Finbarr
    23. May 2009 at 11:00

    I favor policies to sharply raise the US saving rate. That would address the problem you mention.

    No, that would exacerbate the problem (assuming the U.S. saves in dollars). Increased demand for dollars means dollar appreciation, putting an even harder squeeze on domestic manufacturers and exporters.

    What the U.S. needs to do is either a) simply ban China from buying treasuries or dollars or b) buy an equivalent amount of yuan. The goal is to have no net outflow of dollars.

    The problem with Spain was that because they could simply export gold in return for manufactured goods from Britain, it had no incentive to build an industrial base.

    Likewise, the U.S. exports dollars in return for cheap manufactured goods from Asia. In the short run this is good for us, as we get cheap goods. But in the medium and long run the domestic manufacturing base erodes, which is absolutely disastrous.

  4. Gravatar of Bill Stepp Bill Stepp
    23. May 2009 at 15:20

    Indeed if President Roosevelt had not made the mistake of introducing a high wage policy in July 1933, the Depression might have ended fairly quickly.

    Didn’t Hoover have a high wage policy in place when he was president? Was FDR’s wage policy even more adverse to letting the labor market clear?
    Would the Depression have ended fairly quickly even with FDR’s tax hike?

  5. Gravatar of Nick Rowe Nick Rowe
    23. May 2009 at 15:54

    There is one difference between today and the 1930’s: the use of gold as an international reserve currency. In the 1930’s, if all countries devalued against gold, that would cause an increase in the real value of the stock of international gold reserves, the uber-monetary base. That would mean a loosening of world monetary policy, and be good for the world. So national self-interest and collective global interests were aligned (despite the misguided arguments about “beggar-thy-neighbour” devaluations.

    So I can see, in principle, that even a single large country devaluing against gold might be beneficial both for that country, and for the rest of the world (if income effects are bigger than substitution effects).

    But I am having difficulty seeing the same thing today, when there is no counterpart to gold as the uber-base.

    But very interesting post Scott, as always.

    By the way, as I understand it, China’s large savings rate is not so much private savings as public savings. The central bank acts as the government’s agent in creating those public savings, by purchasing foreign assets. The Bank of China’s increasing reserves, and the high national savings rate, are but two sides of the same coin. This does not invalidate your argument; it’s just another way of describing it.

  6. Gravatar of ssumner ssumner
    23. May 2009 at 17:36

    Devin, If the US saving rate rose to equal the rate of investment, then the US current account would balance. So I’m not quite sure what your point is. Generally more saving reduces interest rates, which makes a currency depreciate. I think you may have confused demand for money with demand for bonds.

    Bill, Hoover also had a high wage policy, but FDR’s policy was far worse. The tax increase certainly hurt, but not that much. Taxes were very high in the 1950s-60s. The main causes of the Great Depression were deflation and the NIRA/Wagner Act.

    Nick, I see your point about gold, but I think if you take a closer look it isn’t that different. Let’s suppose you think the monetary base matters in a depression. The 1933 dollar devaluation essentially increased the world supply of gold in dollar terms. And as you say, gold is a quasi-base. China could devalue its currency by increasing it’s monetary base (or reducing demand for base money, which is essentially the same.) This raises the world monetary base, and thus raises world prices. Even better, if other countries “retaliate” and get into a set of competitive devaluations” then they all start increasing their monetary bases, and all start devaluing their currencies, not against each other obviously, but against goods and services. The very respected international economist Barry Eichengreen made the same point a few weeks back, so it’s not just me being wacky. China by itself wouldn’t increase the world’s monetary base all that much, but every long journey must begin with a first step. In the reverse direction, recall that when China had deflation in the late 1990s, it wasn’t because they wanted to, it was because the currency peg made their money supply endogenous. Even with all those reserves.

    My impression is that the Chinese people also have extremely high savings rates (partly because of a lack of safety net.) But I don’t have the exact data. You are right that the government also saves a lot.

  7. Gravatar of Devin Finbarr Devin Finbarr
    23. May 2009 at 18:03


    By saving I thought you meant hoarding the medium of exchange. If people increase their desire to trade dollars for bonds or equities, then the dollar will indeed depreciate. Competitive devaluation is one option for balancing the current account. But it would probably require substantially more monetary inflation than 5% a year, so it would blow away your NGDP target.

  8. Gravatar of Nick Rowe Nick Rowe
    24. May 2009 at 02:56

    Scott: Good! I see your point now, and it makes sense.

    BUT, there are 2 ways China could devalue:

    1. By increasing the money base (what you have in mind).

    2. By increasing national savings (what you don’t have in mind).

    Now, there are 3 ways to increase national savings:

    1. Increase private savings. (Households buy financial assets).

    2. Government raises normal taxes or cuts spending, and runs a bigger surplus. (Government buys financial assets).

    3. Central bank takes a bigger cut of foreign exchange transactions and uses the proceeds to build up foreign exchange reserves.

    That last one is like a tax. You have different exchange rates for different transactions. It’s how Cuba runs its forex operations, with one exchange rate for sugar exports and a different one for tourism, except the proceeds are handed to the government rather than to build up foreign exchange reserves (I used to teach in Cuba).

    I wish I understood 3 better, both theoretically, and how the Bank of China operates in practice. But I don’t, and I can’t get my head around it. But in principle, part of the national savings can be done by the central bank, and that looks exactly like what the Bank of China has been doing.

    So when China hears your call to devalue, they might not hear “increase the money base”; they might hear instead “increase national savings by sterilised buying of more foreign exchange reserves”.

  9. Gravatar of 123 123
    24. May 2009 at 03:20

    China follows market distorting policies that increase Chinese savings rate.

    “Devin, I favor policies to sharply raise the US saving rate.”
    Like abolishing credit cards?

  10. Gravatar of anon anon
    24. May 2009 at 04:51

    China’s central bank intervention changes the form, but not the level of national saving.

    Increasing the monetary base is inflationary / devaluing.

    Sterilizing an increase in the monetary base is deflationary / revaluing.

    Increasing national saving is deflationary / revaluing.

  11. Gravatar of David Pearson David Pearson
    24. May 2009 at 05:31

    A Chinese devaluation would likely affect import prices much more than export volumes. The increase in nominal global AD would be felt primarily as a rise in Chinese inflation, all else equal. Sure, productivity growth would stem some of the inflation, but “all else equal” means that effect is irrelevant as it would have occurred otherwise.

    Rising Chinese import prices would negatively impact the Chinese trade surplus. This, in turn, would reduce available reserves for purchasing U.S. sovereign debt. Although the percentage of U.S. sovereign issuance potentially funded by Chinese reserve growth is plunging (from 100%+ to around 20%), this would still have an impact on U.S. long term interest rates. Higher U.S. long bond yields would in turn force the Fed to increase its monetization of our deficit. Result: higher U.S. inflation.

    You may believe the above dynamic — which increases nominal global AD via inflation — is positive. I would argue that you will soon become acquainted with the phrase “sovereign risk premium”. It is a phrase quite familiar to any emerging markets bond analyst, but rarely attached to words “U.S. Treasury yields”. The process of building in a sovereign risk premium has just begun. When the market doubts that the U.S. can finance its STRUCTURAL deficits without the help of the Fed, then rates will go much higher.

    In the end, QE advocates make two critical assumptions: 1) we will have a V-shaped recover in REAL AD which reduces our structural fiscal deficit and public sector borrowing requirement; and 2) inflation resulting from (temporary) monetization is not a threat as long as we have an output gap.

    Read Kohn’s Princeton speech. You will find the fingerprints of assumption 2) all over it. Without that assumption, his exit plan, his expectation for future inflation being low, and his willingness to keep financing fiscal deficits even as the economy recovers — these all crumble.

    The enemy of the Fed is the sovereign risk premium. The bond market has its own ideas about what price to charge for financing U.S. deficit under a threat of constant monetization.

  12. Gravatar of ssumner ssumner
    24. May 2009 at 06:16

    Nick, I mostly agree. The only thing I’d add is that I think the monetary base assumption is a bit more plausible than many people like to assume. You might think that with their huge international reserves China is sort of like a fortress, and that it’s monetary policy is exogenous. But look how they went from deflation in the late 1990s to too much inflation by 2005. It certainly looks to me like their exchange rate policy makes their money supply somewhat endogenous, even with all those international reserves. Their real exchange rate had to fall in the late 1990s because of the E. Asia crisis, and had to rise in 2005 because of Balassa-Samuelson effect. The government seemed powerless to prevent that from affecting domestic AD and prices.

    123, I think much of their high saving is due to a lack of safety net, which you could argue is a free-market policy. But you are right that the government also saves quite a bit. I do favor distortionary policies along the lines of Singapore, which requires its citizens to put 33% of wages into various fully-funded personal accounts (retirement, unemployment, health care, etc.) This allowed Singapore to abolish the vast majority of their welfare state and still provide universal health care and other programs. It seems to me that Singapore is the most successful economy in the world, although I have to admit I am not an expert, and I know their economy does have some flaws, like an intrusive state, and too much public housing. But for an economist living in the US, their high saving rate looks pretty good.
    I explain my deviation from laissez-faire in the following way:

    1. Force saving is a distortion, but a much smaller distortion than the welfare state.

    2. Laissez-faire policies toward saving may be impossible because of the market failure called “compassion”. If we won’t let those who didn’t save suffer too much, and they know that, then there is a sort of game theory problem. Forced saving is one solution.

    David, A lot of interesting arguments, but I come back to a few basics. The deficit problem we face is a product of the steep recession. Period. A tight money policy caused NGDP to fall sharply and that led to massive bank bailouts, stimulus packages, lower tax revenues, etc. If we reverse the process with an easy money policy, then the side effects will also reverse.
    A tight money policy would just make the deficit so large we will have trouble servicing it, as you say. I am trying to prevent that from happening with an easier money policy.

    Inflation can be kept under control as long as we carefully monitor the TIPS market. Inflation is the last of my concerns. My concern is too little inflation and as a result too much government borrowing, meaning much higher taxes in the future. It’s not a zero sum game, any policy that reduces RGDP makes all other economic problems much harder to solve. Right now we have policies that reduce RGDP, and are paying the price.

    Yes, I believe that since the big drop in NGDP growth caused the worsening of the banking crisis and the steep drop in RGDP growth, if we now have rapid NGDP growth, we will have rapid RGDP growth.

  13. Gravatar of Bill Stepp Bill Stepp
    24. May 2009 at 06:39

    Bill, Hoover also had a high wage policy, but FDR’s policy was far worse. The tax increase certainly hurt, but not that much. Taxes were very high in the 1950s-60s. The main causes of the Great Depression were deflation and the NIRA/Wagner Act.

    Yes, and exacerbated by all those wonderful state banking regs, which disallowed branch banking and prevented bank portfolio diversification. Hail progressive geniuses!

  14. Gravatar of Richard Warfield Richard Warfield
    24. May 2009 at 07:14

    Like many other economists, I believe the trade deficit, or more precisely the current account deficit, simply reflects differential savings rates between the US and East Asia.


    I’m afraid that many economists are mistaking an accounting identity for a clever insight. X-M = S-I, but that equation doesn’t necessarily capture causality. In the case of China, it almost certainly does not.

    Net exports, interest rates, and the savings rate are clearly related to one another in a general equilibrium. There are multiple possible paths of causality. A rise in the savings rate can cause interest rates to decline, which causes the real exchange rate to decline, which causes net exports to rise. Or a fall in the real exchange rate can cause net exports to rise, causing interest rates to fall.

    So in the case of the China-US trade imbalance, which is the chicken and which is the egg? I think the evidence is overwhelming that the path of causality starts with China’s exchange rate policy and related policies that drive up China’s savings rate.

    First, we know that China fixes its nominal exchange rate against the U.S. dollar. Repeat after me: price fixing is prima facie evidence of a market in disequilibrium. I don’t see much need to look for clever alternative explanations for the imbalances when we have an obvious force preventing the most straightforward mechanism to bring about adjustment.

    One may point out that China fixes only the nominal exchange rate, but the real exchange rate is free to adjust through inflation differentials. But with sticky prices, and with the headwinds of the Balassa-Samuelson effect working against the adjustment, the adjustment may not occur in an acceptable time frame. I don’t doubt that in the long run the real exchange rate would reach equilbrium, but in the long run we’re all dead (and the autopsy report shall show evidence of multiple financial crises and trade wars).

    Second, the the savings rate in each country is so pathologically bizarre that it beggars believe that it reflects “naturally” different preferences of Americans and Chinese. Up until recently the savings rates were roughly 50% in China and 0% in the U.S. Neither of these numbers are anywhere close to normal.

    China has a lot of policies that drive up its savings rate. Contrary to popular belief, most of the rise in China’s savings rate was not a result of precautionary saving by households, but rather resulted from high levels of corporate retained earnings. For the state-owned sector (i.e. most of the economy), the decision to retain earnings rather than paying dividends or higher wages is essentially a public policy decision. And for all export industries, the undervalued exchange rate is a sort of direct subsidy to profits.

  15. Gravatar of Joe Calhoun Joe Calhoun
    24. May 2009 at 07:15


    What do you think of increasing the savings rate by reducing taxes on labor (income tax) and capital (corporate tax, capital gains) and replacing it with some type of consumption tax (VAT?).

    I blogged a number of times over the last few years (I’m too lazy to dig out the links) that the politicians threatening China with trade sanctions if they didn’t revalue the Yuan higher should be careful what they wish for. It seems to me that’s like someone saying, “I’ve got a problem and here’s what you need to do to fix it.”

    Also, what do you make of China’s recent move to use their reserves to buy more commodities? Specifically, they seem to be stockpiling copper and oil and also increasing their purchases of gold.

    Finally, what did you think of the markets last week? The dollar seems to have resumed its fall, gold and other commodities are rising and long term bonds got whacked last week. TIPS spreads remained pretty stable. If I’ve been reading your blog correctly, all of this should have been good news to you….

  16. Gravatar of David Pearson David Pearson
    24. May 2009 at 09:00


    Its not what you believe that counts, its how the markets will act. If market participants elect to hedge against future monetization, this would increase velocity. The increase would show up mostly as a rise in the gold price, and it would feed into other commodity prices as well.

    So let’s say that you are making three assumptions:
    1) inflation leads to real growth and a V-shaped recovery.
    2) the output gap leaves the Fed plenty of time to exit.
    3) the markets will agree with 1) and 2) and elect not to hedge against inflation.

    Good luck with all of those, but especially number 3). As with any economic modeling exercise, in the end, its all about the difficulty of forecasting “animal spirits” (i.e. velocity). Economists like to think this is just a “constant”, when in reality it can be highly variable with a material coefficient. That is certainly true here.

    Scott, I would argue that whether your prediction is right or wrong is not what matters. Its whether you are realistic in estimating the probable range of outcomes. Here I really have an issue with your analysis…

  17. Gravatar of Bill Woolsey Bill Woolsey
    24. May 2009 at 10:17


    Where you may be correct about what various people are thinking, the market process you describe raises real interest rates, not inflation. The policy decision that you just ignore is that government must choose whether to inflate away the debt, raise taxes, or cut other spending to pay the added interet burden of the national debt.

  18. Gravatar of Bill Woolsey Bill Woolsey
    24. May 2009 at 10:24


    I wish you would give up on your inflationist rhetoric. What did you say? Your least worry is about inflation being too high. You worry about inflation being too low?

    I worry about nominal income being too low, and I worry about deflaton. I don’t worry about inflation being too low.

  19. Gravatar of TGGP TGGP
    24. May 2009 at 10:35

    if we now have rapid NGDP growth, we will have rapid RGDP growth.
    Like Zimbabwe? I’d like to hear when inflation is or isn’t necessary/sufficient for RGDP growth.

  20. Gravatar of David Pearson David Pearson
    24. May 2009 at 11:45


    A collective move by the market to hold short-duration assets or commodities in lieu of long duration assets is inflationary. It happens time and time again in emerging markets that are facing the threat of a partial default of monetization of fiscal deficits/debt.

    Let’s take the case where private savers overwhelmingly crowd into the very short end of the Treasury curve to minimize inflation risk. That causes demand for long-dated Treasuries to crash and rates to rise. All else equal, the rising cost of financing the debt will cause the fiscal deficit to expand, which in turn will raise forecasts of future total debt and deficits. The higher those deficit projections go, the more the market will believe they will be financed by the Federal Reserve, resulting in permanent increases in the price level. To hedge against inflation, market participants will, on the margin, elect to hold less currency and more hard assets or purchased goods, which drives up inflation. Higher inflation expectations further increase nominal interest rates, etc, etc.

    You are right in one sense. The initial increase in sovereign rates may not be inflationary. What is inflationary is the expectation that the government will turn to the Central Bank to finance structural deficit.

    Again, this is not conjecture, but the historical experience of dozens of emerging market economies.

  21. Gravatar of Joe Calhoun Joe Calhoun
    24. May 2009 at 16:37

    At the risk of making a fool of myself, I’d like to add to Mr. Pearson’s comments. As I’ve said in these comments before, I approach these things solely from a market standpoint. That’s my job and frankly the theory behind why something is happening is much less important than that it is. Economic models may work fine in the lab, but in the real world, all things aren’t equal.

    The risks that Mr. Pearson identifies with respect to the expectations of the market are starting to be realized now. China is taking steps to minimize their use of dollars. They’ve already completed a currency swap with Argentina and are now negotiating with Brazil to conduct trade in Reals and Renminbi rather than routing trade through the dollar. They are also reducing the duration of their portfolio of Treasuries, buying commodities (including gold) and also allowing companies with foreign exchange earnings more flexibility. These companies will be allowed to invest more easily abroad rather than just exchanging their newly earned dollars for renminbi. Monetizing the debt would seem to have some real world consequences.

    The Federal Reserve does not lack for sellers of long dated Treasuries. The last coupon pass saw $47.5 billion offered for a coupon pass where the Fed bought a mere $7.4 billion (maturities from 2013 to 2016). Meanwhile, commodities are starting to rise. Gold got a lot of attention last week, but a wide variety of commodities are on the rise from copper to coffee. Investors are doing exactly what Mr. Pearson states; they are buying short term fixed income instruments and commodities to avoid the effects of the Fed’s inflation.

    Are investors hedging against future monetization? I deal with some Latin American clients and one of them recently made a comment that bears on this situation. He said, “Did you notice that the amount the Fed is devoting to buying Treasuries and such is the same as the amount for the expected deficit?” Alright, I’m paraphrasing but the meaning is clear. As he put it, “I’ve seen that before” (he’s Brazilian). And this particular client knows what that means. Buy real things and don’t tie your money up for very long because interest rates are probably going up. And hedge the currency risk. Or more broadly, invest elsewhere.

    Latin Americans also don’t have a clue about the “output gap”. They know that inflation is not just the companion of economic growth but more often is the handmaiden of recession.

    I’ve been reading here for a few months now and of course Scott is correct that Fed inflation will raise NGDP, but this does not, in my mind, make it necessarily a desirable course of action. As Mr. Woolsey points out, the government does need to choose how to pay for this exercise in profligacy and it appears that inflation is the chosen path. So be it, but let’s not pretend its a good idea. To me, its just a form of repudiation and is basically dishonest.

    The Fed’s actions, in my world, trigger a response. The response when the dollar is falling is to buy real things or stocks in countries that produce real things or currencies of countries that produce real things (Think Brazil, Chile,Australia, Canada). Up until about 2007, it meant buy real estate too. In other words, it was Fed inflation in the early part of the decade that got us in this mess. While I trust they can raise NGDP by inflating, there will be consequences for those actions this time just as there were last time. More of my (and my client’s) assets will be invested outside the US and in commodities. It’s what the market tells me to do and I don’t argue with the market. If others see things the way I do (and it appears they are doing so more every day), then that can only mean less capital for productive US companies and with the way things are today, that would seem a less than ideal outcome.

    What the Fed needs to do is stabilize the dollar. Not necessariy right here, but at some point they have to stabilize the currency. The volatility of commodities is a direct reflection of the volatility of the currency. The consequences of that are immense. How many companies lost money last year (or even filed bankruptucy) trying to hedge the volatility of either an exchange rate or a commodity price? Capital devoted to managing the risks associated with currency volatility is capital than can’t be devoted to a more productive activity. All this hysteria over derivatives is also a direct consequence of currency volatility. How many interest rate swaps would be outstanding if interest rates weren’t as volatile? What would the volume be on the CME if commodities had a standard deviation of 7 instead of 18?

    I spend an inordinate amount of my time watching the currency and commodity markets. It would be nice if I could spend that time, as I once did, researching companies and making long term investments. Unfortunately, that isn’t the world I occupy at the moment. The Fed has turned me into a speculator. That is the only way to survive in such a volatile environment. So yes, Scott, I think you’re right about the long term if the Fed were to target NGDP all the time. We’d get smoother growth patterns and reasonably low inflation. But from here, initially, the results are likely to be inflation and a cheaper dollar. To me, that is rational and something I can live with if they were to change how they conduct monetary policy permanently. I don’t however, expect that.

  22. Gravatar of Lawton Lawton
    24. May 2009 at 20:01

    @Joe Calhoun: do you have a favorite link arguing in favor of stabilizing the dollar and what that means in practice? For example, I’ve seen interesting plots of Taylor Rule vs. Greenspan’s actual policies; is there an equivalent for a “stable” dollar? A quick Google search turns up a Martin Feldstein paper against stabilization. (And plenty of other links of course.)

  23. Gravatar of Joe Calhoun Joe Calhoun
    24. May 2009 at 21:43


    If the goal is to reduce volatility (and I think whatever the models used, that is what everyone is really after) it seems to me that the first step is to reduce the volatility of what we can. Stabilizing the value of the currency will reduce commodity price volatility and that has real benefits. There are two studies I know of that demonstrate this well:


    By the way, I suspect that interest rate volatility would be lower if the currency were stable, but I haven’t read the research on the subject so I can’t back that up with data.

    I think the role of a central bank should be limited to what it can control and the one thing I see that they have almost absolute control over is the exchange rate. Managing the supply of dollars to match demand is still not an easy task, but seems much easier than managing inflation and growth as the Fed is currently tasked. I don’t believe we can create real growth by printing more dollars. Was the growth we created from 2002 to 2007 real growth? Or did we just create wasteful economic activity? To my thinking it is all the policies other than monetary policy that most affect the long term growth potential of an economy.

    I think Scott’s approach of targeting NGDP would probably produce a stable currency in the long run. The only problem I see is accounting for changes in productivity (which I see as being driven by all those other policies). Changes in productivity would seem to change the non inflationary NGDP target. But, hey I’m not an economist so maybe someone more qualified can comment on that.

    I think the point is that we need consistency in monetary policy. Whether we get that by targeting NGDP or the value of the dollar I think might be irrelevant.

  24. Gravatar of Joe Calhoun Joe Calhoun
    24. May 2009 at 21:52

    One more thing. I see capitalism as inherently deflationary. Capitalism is afer all about becoming more efficient and productive. If that is the case, don’t we waste some of the benefits of capitalism if we always maintain an inflationary monetary regime? So my ideal monetary policy would maintain that deflationary bias so that the benefits of capitalism were more evenly distributed. Inflation exacerbates wealth and income inequality because the wealthy have the means to protect themselves from it while the poor do not.

  25. Gravatar of Bill Woolsey Bill Woolsey
    25. May 2009 at 03:19


    Crowding into the short term end of the market lowers real and nominal interest rates. It doesn’t cause the fiscal deficit to explode. The increase in the yields on long term treasuries doesn’t impact the treasury at all. It imposes capital losses on current investors. New borrowing by the Treasury will have to be short. It will actually be cheaper.

    The increase in the demand for treasuries is inflationary. And, of course, if that happens, then the problem of inadquate nominal expenditure is solved. And it is time to reverse quantitaitve easing.

    If you have a government that is already taxing all it can, and has borrowed all it can. And it creating money to fund politically necessary expenditure, then this crisis can occur.

    The U.S. can cut spending. And it can raise taxes. The only question is whether U.S. politicians prefer that to monetizing the debt.

    If bond investors believe that QE isn’t aimed at maintaining nominal income, but rather with funding government expenditure with no thought for what happens to nominal income, and, more importantly, with the implication that nominal income will rise past it previous growth path and to new growth rates generated by budgetary needs, then this will happen.

    If that does happen, once the growth path of nominal income is reached, QE must cease and be reversed. If the bond investors are suddenly irrational and don’t see this (unlike seeing the QE before and wrongly interpreting it as the first step to hyperinflation) then the interest cost of funding the national debt will rise (but this happens when the short term rates rise and rolling over existing ones as they come due now requires higher rates.) And then this will require other cuts in government spending or tax hikes.

    QE isn’t a problem. The huge budget deficits are a problem. But the U.S. is far from a sistuation where they are trully unsustainable. Taxes could go up a lot in the U.S., and government spending could drop a lot.

  26. Gravatar of David Pearson David Pearson
    25. May 2009 at 04:34


    Again, there is what you believe, and then there is what the markets will expect; I’d put more importance on the latter.

    You propose that the American polity will stomach deficit contraction amidst high unemployment, state fiscal crises, etc. Please note that the particular problem of this recession will increasingly be not just unemployment, but the DURATION of unemployment: these long-term unemployed will come to depend on government assistance for subsistence.

    So the whole point is that the markets may expect that the government, or polity, will favor monetization over deficit contraction. This is a POLITICAL, not economic, calculation, and the markets make it all the time in emerging economy sovereign debt.

    If the markets disagree with you, velocity will rise, inflation will rise, the dollar will devalue, short term debt roll-overs may become problematic, and inflation, the risk premium, and the possibility of partial default will all enter in the market’s thinking.

    Please don’t confuse your point estimate with the possible range of outcomes. This is a dynamic, non-linear process, and small changes in expectations can lead to large changes in outcomes. This is something that QE enthusiasts absolutely need to ignore in order to make their case.

  27. Gravatar of ssumner ssumner
    25. May 2009 at 04:51

    Bill Stepp,

    I agree, State branching restrictions led to bank failures, which led to currency hoarding, which led to deflation.

    Richard, Obviously I meant “caused by” when I said “reflects.”

    I don’t agree that nominal exchange rate rigidity causes long run deficits. You are right about short run price stickiness, but surely that lasts a few years at most, and can hardly explain our deficit with China. China was experiencing deflation in the late 1990s, and they didn’t want the deflation. It was being forced on them by their exchange rate peg. So the yuan was certainly not undervalued then. Yes, the equilibrium real exchange rate has since risen, but by your argument of slow adjustment, China should still have the same overvalued yuan they had in the late 1990s. By 2005 they had rising inflation, again the market adjusting the real exchange rate to its proper position. The bottom line is that long run economic problems can never be explained by short run price stickiness.

    There is debate about their actual saving rate. But if it is high, then more power too them—we should copy their public policies encouraging saving, as I mentioned in response to another comment.

    Joe, Yes, get rid of the corporate and personal income taxes and go with a payroll and VAT tax system.

    TIPS spreads have been widening since March, so that is good news. The bad news is that expected NGDP growth is still very low. Too low for normal times, and way to low for 9% unemployment (which is going higher, despite all the people telling me I should worry more about inflation. Lots of people worried about inflation in the 1930s.)

    David, Inflation expectations are observable in real time, so I am not worried about inflation sneaking up on us. If expected inflation got high, then I would be worried. (Although NGDP growth is the better indicator.) I do not use the Keynesian “output gap” theory of inflation. I understand that inflation can be high in a recession. My point is that I don’t expect the rate of inflation to sharply accelerate when unemployment is very high and rising fast. Even in the 1970s that never happened (although many people think it did.)

    Bill W., Yes, I should stick to NGDP, but its hard when people keep asking me about inflation. Right now they are both too low. So even inflation targeters like Bernanke should be worried.

    TGGP, My assumption is that if we had 5% NGDP growth, we wouldn’t have 10% inflation and minus 5% real growth. I can’t prove it, but I know it’s true.

    David#2 Our national debt problem isn’t worsened by inflation (check out the 1970s) it is worsened by deflation. In real terms inflation makes the debt smaller.
    We are not an emerging market. When they have financial crises their currencies fall every time. The dollar has risen against the euro over the past 12 months—by quite a bit. We are not a banana republic, but if we continue with our contractionary monetary policies, we could go the road of Argentina. For those who don’t know, Argentina followed deflationary policies on the advice of conservatives (a stable exchange rate to be precise) and end up with such a bad depression that the leftists took over and tore up property rights. Whose to blame? The conservatives. Ditto for the Republicans in the early 1930s.

    Joe, If you have been reading my blog for a few months then you know I am not an inflationist. In the long run I want about 2% inflation, and in the short run my 5% NGDP rule would lead to less than 2% inflation, because the SRAS is very flat at high unemployment rates.
    I am also not worried about the Fed creating a lot of inflation. If investors were really worried about inflation, then commodity prices wouldn’t be half their level of a year ago, and TIPS would be a very popular investment.

    I don’t see how 2% inflation hurts the poor. The Fisher effect insures inflation is priced into nominal rates. Inflation does impose a tax on capital (because of nominal interest taxation), but that would hurt the rich at least as much.

    Bill W. Yes!! Big deficits are the real problem. What so many conservatives don’t see is that the big deficits we have didn’t just happen, they were caused by falling NGDP. Not partly caused–but rather totally caused by falling NGDP. If they want to get rid of big deficits–which some people claim threaten to turn us into a banana republic, then they should support getting NGDP growth back to the normal 5%. I just don’t see why that is so complicated.

  28. Gravatar of David Pearson David Pearson
    25. May 2009 at 06:03


    Features of (Latin) emerging markets:

    -high C.A. deficits
    -chronically low private savings
    -credit crises produce bail-outs that cost a multiple of GDP
    -PSBR is extremely high relative to private domestic savings
    -Central Banks are willing to monetize structural deficits
    -sovereign risk premium rises when GDP forecasts fall, which leads to spiraling debt costs and deficits
    -resulting flight capital drives large devaluations

    All of the above apply to the U.S., except the last two. Well, those Treasury yields sure have been misbehaving lately, so maybe its just that last one we’ll NEVER have. Right?

  29. Gravatar of Joe Calhoun Joe Calhoun
    25. May 2009 at 06:57


    I know you’re not an inflationist and I’ve said as much in my other comments. I believe your model would result in low inflation in the long run, but we have to get to the long run. Are you saying that if oil prices return to $150/barrel to achieve your 5% NGDP, that isn’t a problem? If corn prices return to $8/bushel that isn’t a problem? If natural gas prices return to $15/mcf that isn’t a problem? Returning to the status quo of last July would not in my opinion be a good outcome. And that seems to be where we are headed.

    Maybe the Fed will be able to extract themselves from the markets in a timely fashion. Maybe the Fed will time the shrinking of its balance sheet perfectly. Maybe the politicians will reduce the budget deficit at the appropriate time. Or maybe as you suggest, the deficit will fall naturally as NGDP returns to 5%. Certainly higher growth and tax receipts will reduce the deficit some but will we get back to 1 or 2% of GDP? Or heaven forbid, a balanced budget? There is nothing in my experience to suggest that those outcomes are likely. I think it is much more likely that the Fed stays at the party too long and the spending programs enacted by Congress in this “emergency” become permanent. And the structural deficit becomes larger.

    How does the Fisher effect protect the poor from inflation? Isn’t the Fisher effect concerned with interest rates? Aren’t nominal wages sticky in both directions? Do you really believe that when gas prices were over $4/gallon that the poor weren’t hurt any more than the rich? The wealthy are more protected from inflation because they own assets which benefit from inflation. When the Fed inflates and house prices rise, those who already own property benefit first. The poor may not be hurt by this initially if rents don’t rise, but the wealth gap becomes wider because they don’t own any assets that benefit from inflation. I’m suggesting that we use the stickiness of wages to benefit the poor in the long run by running a mildly deflationary monetary policy. Unlike most people, I don’t fear deflation if it is deflation due to rising productivity. I don’t buy the argument that deflation is always a bad thing.

    I was led to this website by a search for more information on Earl Thompson’s work and I’ve stuck around because I like the idea of a predictable monetary policy. To me that is the key to achieving more stability. But I also think that, unlike Thompson’s labor standard, we can’t just adopt your system and expect no transition effects on growth and inflation. Getting from here to there will involve some tradeoffs and I suspect the tradeoff in the current condition is higher inflation than you advocate. Since I’m not an economist, I don’t think I have the ability to express the reasons for that in your language, but it is the outcome I and a lot of others in the market expect. We may be wrong, but in the interim, expectations matter.

  30. Gravatar of Thruth Thruth
    25. May 2009 at 12:20

    David P: “-credit crises produce bail-outs that cost a multiple of GDP”

    I think you might be stretching a little bit. Last I checked we weren’t anywhere near $15tr.

    I don’t think anyone, including Scott, is denying the possibility that inflation could take off sometime in the future. It seems to me that Scott just sees it as both remote and something that the Fed should respond to when the data indicates it’s happening. The Fed can only take one monetary policy stance (though using a variety of instruments) and they have to balance both inflationary and deflationary risks. The data at this stage suggest they are erring on the side of deflationary.

    I think I’m siding also with Scott on whether the Fed can respond to rising future inflation with its present balance sheet. If velocity were to suddenly take off, it’s not clear to me why the Fed balance sheet even matters (especially wrt Treasury security holdings — that surely has to be zero sum). Why does the Fed have to unwind the balance sheet. Can’t they simply raise short rates as they normally do and just let the portfolio wind down of its own accord? Does it really matter if the Fed takes losses? At least with respect to holdings of Treasuries, it’s a zero sum game between the Fed&Treasury (don’t forget that the Fed gains and losses are included in the deficit). For other assets, the losses are most likely small and will probably just result in reduced Fed dividends to Treasury. Perhaps the argument is that the Fed will lose its independence if it takes big losses? I think there’s more to fear about the Fed losing it’s independence in the face of big fiscal deficits, and as Scott has said, tight money and deflation will get you there.

  31. Gravatar of ssumner ssumner
    26. May 2009 at 04:08

    David, As I said, it is possible that the US would become a banana republic if we continued down this road. That’s why I want changes in policy. Both the markets and I think the outcome you worry about is very unlikely. Our political system is far too orthodox to make such an outcome likely.

    Joe, I believe that last July was an absolute paradise compared to our current situation–which is still deteriorating rapidly. And I think most American’s agree. I am pretty sure that most Americans think the economy has gotten much worse. I even saw a monthly “happiness survey” of Americans that dropped off significantly last fall. High commodity prices are the result of worldwide prosperity, hundreds of millions of people moving out of abject poverty. You bet I want to see those days again. And for those who don’t care about the world’s poor, how about the stock market since July? How about the US unemployment rate?

    Your sticky wage plan with deflation won’t work, as wages are flexible in the long run. Japan has been doing your proposed monetary policy for the past 15 years, BTW. They’ve had pretty steady, but mild, deflation.

    If we have 5% NGDP growth, I think we’d get at least 3% real growth in the short run, so I am not worried about inflation. In the long run the 5% number might need to be reduced slightly, as the trend growth rate may fall to 2-2.5%.

    Thruth, Very well put.

Leave a Reply