A Theory of Non-Relativity

Yesterday’s online version of The Economist, also known as known as Free Exchange, did a nice piece on my recent China post.  They contrasted my views with those of Paul Krugman, and also asked a few questions.  Here I’ll try to respond to those questions, but first I’ll clarify exactly where Krugman and I differ.

First some areas where we agree.  Both Krugman and I see the biggest macro problem over the past 14 months as being the sharp fall in aggregate demand over most of the world.  We are both strong supporters of additional stimulus; indeed we are both among the diminishing band of economists who still favor additional stimulus.  We both strongly oppose monetary tightening by the Fed, ECB, and BOJ.  But there are also some key differences, which may have indirectly contributed to our opposing views on the Chinese Rmb.  In the US I favor additional monetary stimulus and oppose fiscal stimulus.  Krugman is not opposed to additional monetary stimulus, but doubts the effectiveness of such actions when the policy rate is near the zero bound.

There is pretty general agreement that central banks have almost unlimited ability to depreciate their currency.  And most major central banks (including the Fed and the People’s Bank of China) have the resources necessary to sharply increase their nominal exchange rates, if they so desire.  So I agree that the Chinese government could appreciate the yuan in nominal terms.  Let’s put off the issue of saving rates and current account deficits for the moment, and grant Krugman’s assumption that the PBOC could also appreciate the real Chinese exchange rate, at least for a considerable period of time.  The question is whether it would be wise.  I think it would be bad for China, and bad for the world.

However, my reasons are not what you might expect, as I look at the issue from a money/macro perspective, not a trade perspective.  If the Fed were to adopt a much more aggressively expansionary monetary policy, and if this led to a higher value of the Chinese yuan, I’d be in favor.  So why does it matter how the Chinese currency appreciates, as long as it appreciates?  Because an exchange rate is merely the relative value of two currencies, it tells us nothing about the absolute value of those currencies in terms of goods and services, which is much more important.  If the Fed adopted a much more expansionary monetary policy, and if the PBOC kept its policy stance the same, then world monetary policy would become more expansionary, and world aggregate demand would increase.  That would help everyone.

I’m going to go out on a limb and suggest that Paul Krugman very likely agrees with the preceding analysis, at least at a technical level.  So why is he calling on the PBOC to appreciate the yuan against the dollar (a tight money policy), rather than call on the Fed to depreciate the dollar against the yuan.  I think there are at least 4 explanations he might give:

1.  Krugman might argue that we have already depreciated the dollar (against the euro and yen) and the Chinese have refused to play ball.  Rather than keep their exchange rate stable against a basket of currencies, they have let it fall by tying it to the falling dollar.  And I am sympathetic to this argument.  But the bottom line is that no matter how weak the dollar looks, it is not as weak as it needs to be in absolute terms, i.e. against goods and services.  Expected growth in prices and NGDP (my preferred indicator) is still too low.  So yes, the yuan is weak against the euro and yen, but the currency it is stable against (the dollar) is itself too strong.

2.  Krugman might argue: “Yes, it would be better to depreciate the dollar by making US monetary policy more expansionary, but at the zero rate bound that is not possible.”  (Or perhaps he’d argue that it is theoretically possible but would require extreme and risky unconventional policy steps that the Fed is too conservative to take.)  Much of this blog has been devoted to refuting that argument.  I think the Fed could stop paying interest on reserves, perhaps even impose an interest penalty on excess reserves.  And they could engage in quantitative easing (something they have not done during 2009, despite reports to the contrary.)  And by far the most important and powerful step would be to set an explicit nominal target, a trajectory for the price level or NGDP to grow over time.

3.  Krugman might argue that because the Chinese economy is growing vigorously they don’t need any more stimulus, and they are stealing jobs from the rest of the world with a low yuan policy.  There might be a bit of truth to this argument, but I would argue the net effect of their action is much more likely to be positive.  A strong Chinese economy boosts world AD.  Furthermore, other countries are free to adjust their currencies appropriately.  The South Koreans responded to the Chinese policy by letting the won depreciate during the crisis.  (Don’t ask me about the BOJ, I have not understood one thing they’ve done since 1993.)  You might wonder whether this could lead to a currency war, each country trying to depreciate its currency against the others.  Maybe, but that is precisely what you’d like to see happen in a world economy where AD has collapsed because conservative central bankers don’t understand that their policies drive the expected rate of growth in nominal GDP.  Or perhaps I should say they don’t understand that this well established proposition applies not just in normal times, but also during financial crises.  Obviously competitive devaluations could get out of control, leading to excess demand.  But I’m with Krugman on this issue, the threat to the world economy for the next few years is too little AD, not too much.  And the markets seem to agree.  Not just the equity markets, which clearly favor monetary ease, but the indexed bond market, which shows very low inflation expectations over the next two years.  (BTW, if you think only a slightly nutty economist could recommend a round of competitive devaluations, consider this post by the respected Barry Eichengreen.)

4.  Krugman might argue that the Chinese could insulate their domestic economy from the impact of a yuan appreciation.  They could appreciate the yuan without tightening domestic monetary policy.  After all, they still have strong currency controls.  But these controls are not as strong as they seem.  In the late 1990s they were not able to prevent yuan overvaluation from causing Chinese deflation, and in the period leading up to 2005 they were not able to prevent inflows of hot money from driving inflation and asset prices higher.  I am not convinced the Chinese government has firm control over their real exchange rate, no matter how powerful it may look to outsiders.  Admittedly, their foreign exchange reserves have reached formidable levels.  But they are holding these reserves for a reason, and I wouldn’t expect them to easily give them up.  I am most concerned by the possibility that the Chinese government would try to offset a high yuan policy with a policy of easy credit from state-owned banks—which lend almost exclusively to the relative inefficient SOEs.   We have already seen a disturbing shift in the Chinese economy, as the export sector (comprised on smaller private firms) has recently shrunk, and instead the government has forced banks to massively expand their lending to SOEs.  A policy of currency appreciation combined with increased lending from state banks to SOEs would further set back Chinese market reforms.

So far I have focused on only one issue, an issue where Krugman ’s goals are similar to my own—the need for more stimulus, more aggregate demand.  But Krugman also raised a separate and much different issue, the so-called “imbalances” in the world economy—imbalances that were supposedly exposed by the recent crisis.  I just don’t buy this argument.  I’d like nothing more than to go back to the ‘imbalances” of mid-2008; with oil at $140 a barrel, the world economy strong, US unemployment in the mid-5% range, and a huge US current account deficit.

If we do return to prosperity over the next few years, would it be a bad thing if the US current account deficit rose back up to 5% of GDP?  I don’t see why.  Or perhaps it would be more accurate to say that I don’t think either monetary policy, or exchange rate policy should try to “fix” this so-call imbalance.  I would favor a fiscal tax/spending system that was much more pro-saving, something along the lines of Singapore, but that is a different debate.

When I lived in Australia in 1991 it was in a severe recession.  The Aussies I talked to were pessimistic about their future, and pointed to their big current account deficit.  I didn’t see any problem with a current account deficit.  Since 1991 Australia has had arguably the best macroeconomic performance of any major developed economy.  They are the only major developed economy to have had had zero recessions during this period (we’ve had two.)  And yet they have continued to run huge CA deficits year after year, roughly comparable to the US as a share of their economy.  Why did they get off so lightly in this crisis if a big CA deficit is the root cause of our problems?  When will they have their day of reckoning?  I doubt they will ever have a “day of reckoning;” the Australian CA deficit reflects differential patterns of saving and investment between Australia, whose population is growing fast through immigration, and countries like Japan, which has a declining population.  Some might argue that there is one big difference between Australia and the US; the Aussies generally don’t run large budget deficits.  Maybe so, but I don’t think Krugman would want to make this argument, as he thinks the US budget deficit is currently much too small.

Krugman strongly disagrees with those (like me) who believe it is more useful to think of savings/investment patterns driving the current account, rather than exchange rates.  I wouldn’t deny that in the short run the Chinese government could stop accumulating assets and let the yuan appreciate.  And this decline in government saving might reduce the CA deficit in the short run, although history shows that CA deficits often stubbornly refuse to respond to exchange rate fixes.   But if these current account “imbalances” are a problem, they are a long term problem, not a cyclical problem.  And there is little evidence that fiddling with the nominal exchange rate can influence the long run real exchange rate.  If political pressure forced the Chinese government to appreciate the yuan in nominal terms, the gesture would be gradually eroded by offsetting deflation (just as in the late 1990s), as there are very strong reasons for the high savings rates of both the Chinese government and private sector.

Now let’s look at the questions from The Economist:

This is a very interesting take and worth considering. I have some questions, though. First, would appreciation of the renminbi against the dollar constitute general appreciation? There are other currencies in the world. The Wall Street Journal notes that RMB depreciation thanks to its dollar peg has meant appreciation of other Asian currencies, many of which are floating (or “floating”), against the RMB. (The Economist makes a similar point in the latest print edition.) In other words, how does the current world compare with one in which China allows its currency to float against the dollar but also targets a level of nominal GDP around 10% and adjusts monetary policy accordingly? Exchange rate policy may be a form of monetary policy, but that doesn’t mean it’s the optimal form of monetary policy.

Unless I am mistaken, most of the other Asian currencies have depreciated since mid-2008.  The obvious exception is Japan.  It is true that there has been some recent appreciation with the strong recovery in Asia (helped by their earlier depreciation), but I think that is appropriate given the recent differences in the relative performances of the US and the Asian economies (ex-Japan.)  It is interesting to compare Japan and South Korea, two relatively developed economies that both export capital goods to China.  Because the Korean won depreciated during the crisis last year, Korean NGDP hit an all time high in the 2nd quarter of 2009 (and I believe was also extremely strong in the 3rd quarter, just announced.)  At the same time the BOJ allowed the yen to appreciate sharply, and as a result its NGDP was 8% below its 2008 peak by 2009:2.  That’s nominal GDP, in case someone wants to look for “real factors” as an explanation.  Now I obviously would not argue that there is anything like a perfect correlation between changes in exchange rates and changes in NGDP.  In fact, real exchange rates more often appreciate in booming economies.  But Japan and Korea faced essentially the same external shock on 2008, and I think the vast difference in their NGDP growth rates since is at least partly due to the very different exchange rate policies, which in this particular case do reflect different monetary policies.  I’m not certain whether a stronger RMB would lead other Asian countries to appreciate their currencies.  But if it did so, it would magnify the contractionary impact of the Chinese action on world AD.  And there is obviously some tendency for countries to respond to exchange rate policies of their larger neighbors.

Secondly, Mr Sumner suggests that China’s high savings rate is behind its current account surplus, rather than currency policy. Mr Krugman offers some thoughts on the matter here. I don’t know that this is much of a reason to favour or disfavour exchange rate adjustments.

I answered this question in my previous discussion.  To reiterate, I don’t see CA imbalances as being a “problem” (although they may be symptomatic of deeper problems in the economy.)  If Krugman is right that they represent a long term threat to the health of the world economy, then long term structural changes (i.e. encouraging savings in the US) are more likely to be effective than a policy of manipulating the nominal exchange rate.  Monetary policy has little impact on long term real exchange rates.

Is it right to credit China’s currency policy with cutting off the possibility of a deep recession? Mr Sumner has generally argued that monetary policy has been too tight, and as China’s dollar peg looks to him like a particularly substantial effort at monetary easing, it’s logical that he would conclude that the peg has been the decisive factor. I’m not sure most economists would agree. And again, is it the case that the current world is preferable to an alternative one, in which the Chinese allow their currency to float but also pursue a much more aggressive monetary policy?

I’d break this question down into two pieces.  Did the recovery in China that began in March dramatically reduce the chance of a deep depression?  I am convinced that it did.  Not only did US equity markets begin to rally at that time, but the Asian markets often seemed to be leading the way.  There were occasions in the spring where positive stories out of Asia seemed to be the only positive news impacting US markets.  The harder question is what caused the Chinese recovery.  I suppose at first glance the most visible Chinese stimulus was not its exchange rate policy, but rather its fiscal/monetary stimulus package, which largely consisted of loosening regulations on lending by state-owned banks.  This boosted the real estate market, and spending on government infrastructure such as high speed rail projects has also increased impressively.  This could be viewed as a point in Krugman’s favor.  But people underestimate the impact of monetary/exchange rate policies, because the effects often work in unexpected ways.  I mentioned in the earlier post that the US dollar devaluation had a very powerful effect on US real and nominal GDP after March 1933.  But none of this increase was attributable to changes in trade flows.  Instead all of it occurred as rising inflation expectations drove up asset prices and increased domestic aggregate demand.  Krugman has written approvingly of a 2008 AER paper by Eggertsson that made this argument.  Note that the rapid recovery in the spring of 1933 occurred before any significant fiscal stimulus had occurred.  And finally, weekly changes in the WPI and the Dow during 1933 were strongly correlated with dollar depreciation against gold.   If China had appreciated its currency this year then deflationary expectations in China might have become entrenched, threatening industries that are seemingly unrelated to trade, such as housing construction.

In the Krugman post linked two paragraphs above, the author notes that for America to return to full employment and reduce its trade deficit necessarily requires that either America experience deflation, other countries experience inflation, or the dollar depreciate. Mr Krugman says that central banks abroad will not allow inflation and deflation is very painful, so for adjustment to take place exchange rates must change. Another way of putting this might be to say that since central banks won’t allow inflation, the only way to get enough of a monetary policy boost to do any good is for a large economy to allow its currency to depreciate. Perhaps that is what Mr Sumner is saying.

Again, I don’t think exchange rate manipulation can solve these long run problems.  If we get more pro-saving policies in the US, that should lead to the appropriate adjustments in real exchange rates over time.  If all countries target inflation at 2%, then those adjustments will occur through nominal exchange rate changes.  If countries fix their currency to the dollar, then the adjustment will occur through higher inflation in places like China.  In the long run there is really no other way to address this problem.  The final two sentences inappropriately mix up two issues.  I am saying that we need more aggregate demand, and thus more inflation.  Indeed people often forget that inflation is one of the objectives of both monetary and fiscal stimulus.  So currency depreciation is only appropriate if required to get inflation expectations over the next two years up close to 2%.  (Currently the two year inflation forecasts in both the futures and TIPS markets are well below 1%.)  In my view it is the Fed, not the Treasury, which determines the fate of the dollar.  So if the Fed is opposed to boosting inflation up to 2%, then there is not much the Treasury can do in terms of ‘talking down the dollar.”  The Fed ought to be in favor of more stimulus, and indeed some FOMC members have said this.  But others have recently sent out hawkish signals, so policy is hopelessly non-transparent.  Would the Fed like to raise 2-year inflation expectations up from 1/2 % to 2%?  I haven’t the faintest idea.  And that’s not good.

Mr Krugman implicitly argues that a recovery which allows imbalances to persist is no recovery at all. Mr Sumner obviously thinks that exchange rate levels are irrelevant to imbalances. And I suppose I’m struggling to square the two views.

That’s not quite my view. I do think that currency manipulation can have a short run impact on trade imbalances.  But that is not the problem that Krugman was addressing.  Long run changes in savings and investment can lead to changes in the current account, which will generally require an adjustment in the real exchange rate.  So I would never claim that exchange rates are irrelevant, just that they are not an effective policy tool for solving real problems.

Photo: stan