The RMB and the Magic of Accounting Identities
ByMichael Pettis Not only can I write worriedly about rising contingent debt levels three or four years before they become obvious, but I can also revisit a controversy that took place March involving Paul Krugman and Stephen Roach. I revisit this old controversy because although the period of nasty trade dispute seems to have come to an end, with conciliatory noises being made between the major parties, trade tension is not going away. In fact it is going to become internationalized, with more noise coming from other developing countries, who have already begun complaining about RMB policy (for example last week the governors of the central banks of both India and Brazil came out with strong statements about the RMB). In the end, large trade deficits are not politically compatible with high unemployment, and in my opinion things only will get worse on both fronts.
So while this post might not be currently topical, the subject will again become so very soon. To start, and in response to many of the comments from recent postings, especially some emailed comments from friends, I wanted to work through Paul Krugman’s point on currency intervention because, as I see it, he is simply using accounting identities to set out the parameters of how to think about the issue of RMB revaluation.
Accounting identities cannot be violated, neither in practice nor in theory, although depressingly enough much of the discussion of trade balances, even when conducted by economists, brazenly violates these identities. It may make sense then to place the discussion solely in that context so that at least we can all agree where we legitimately disagree. Here is what Krugman says:
Let me start with a proposition: the right way to think about China’s exchange rate is, initially, not to think about the exchange rate. Instead, you should focus on China’s currency intervention, in which the government buys foreign assets and sells domestic assets, on a massive scale.
Although people don’t always think of it this way, what the Chinese government is doing here is engaging in massive capital export – artificially creating a huge deficit in China’s capital account. It’s able to do this in part because capital controls inhibit offsetting private capital inflows; but the key point is that China has a de facto policy of forcing capital flows out of the country.
Now, bear in mind the two basic balance of payments accounting identities:
Capital account + Current account = 0
Current account = Domestic savings – Domestic investment
By creating an artificial capital account deficit, China is, as a matter of arithmetic necessity, creating an artificial current account surplus. And by doing that, it is exporting savings to the rest of the world.
Notice the first identity. China’s current account surplus must be equal to its capital account deficit. It doesn’t matter whether you think China’s trade surplus is caused by policies that force households to subsidize producers, or that place tariffs (hidden or explicit) on imports and subsidies on exports, or that force heavy currency intervention by the PBoC. In the end, these are all one and the same thing, and will have automatic balance of payments consequences. Krugman argues that since China restricts the capital account, the PBoC’s currency intervention automatically results in the Chinese capital account deficit and the current account surplus.
If this true, what are the consequences? Let us for simplicity call the rest of the world the US. If China runs a capital account deficit, the US must run a capital account surplus. That is, it must import savings from the rest of the world. This is one of the inviolable accounting identities.
One country’s surplus is another’s deficit
Since the current account is the obverse of the capital account, this also means that the US must run a current account deficit. Here is where another accounting identity becomes very useful. The total amount of savings the US imports is defined as domestic investment minus domestic savings – in other words by definition the US imports whatever additional savings it needs to fund its domestic investment.
Remember that this is also the same thing as saying that the excess of US investment over US savings is equal to the current account deficit – because the capital account surplus is the same as the current account deficit, right?
So if the US runs a capital account surplus, it must run a current account deficit, and if China runs a capital account deficit, the US must run a capital account surplus. This all nicely balances out because if the US runs a current account deficit, China must run a current account surplus.
Remember, however, that there is no direction of causality implied here. These things must all happen simultaneously, and there is nothing in the accounting identities that tells us which caused the other. If PBoC intervention forces a capital account deficit on China, as Krugman says, then all those other things follow automatically. Alternatively, if Americans decided independently to go on a consumption binge that forced the US into a current account deficit, then all these other things must also follow automatically.
Given the sheer size of PBoC intervention, the tremendous concomitant need to sterilize and repress interest rates, and the resistance to appreciate the currency, it seems pretty clear to me that at least part of the reason for PBoC capital exports was as a policy choice. This is especially likely, I think, since the explosion in reserves at the PBoC and other Asian countries seems to have begun within a few years of the 1997 crisis.
Now if the initial cause was the PBoC capital exports, of course the Fed could possibly have foiled the capital impact of the PBoC intervention by raising interest rates and forcing up unemployment in the US.
This is the point that Martin Wolf has often made, and if the Fed had done so it might have caused private businesses to cut back on their investment more quickly than the resulting decline in savings, and the US could have effectively blocked capital imports. The rising unemployment would have reduced US consumption and US imports, which would have reduced the US current account deficit. Remember, these are just the opposite sides of the same coin and one automatically implies the other. If the excess of investment over savings declines, so does the current account surplus.
For whatever reason, right or wrong, the Fed didn’t do this. The result was that the US had to run a capital account surplus. In the US, however, there are two ways the capital account surplus must resolve itself. Since the capital account surplus is equal to the excess of investment over savings, if the capital account surplus rises, broadly speaking, either savings must decline, or investment must rise (or some appropriate combination of both in which the excess of investment over savings rises).
Jack up investment
Under these conditions it is unlikely that private investment will rise. The US current account deficit means that US demand is shifting abroad, perhaps because currency intervention has made US production less profitable and foreign production more profitable (the overvalued dollar reduces the profitability of US investment and increases the profitability of foreign investment). It is more likely for US businesses to increase investment abroad at the expense of investment at home.
That leaves two other possibilities – either total investment rises because government investment rises, or private savings must decline. The “correct” way for the US to have dealt with the US current account deficit might have been for a sharp increase in government investment, say in infrastructure spending, scientific research, education, and so on. This would have presumably improved the productivity and profitability of US production in the future, so that with lower costs, the US could have eventually regained the edge it lost with the overvaluation of the dollar. The US would have still run a large trade deficit, but this deficit would be the result of a surge in investment rather than a surge in consumption.
But if the US government did not increase investment by enough, the automatic consequence had to be a decline in the US savings rate. There are many ways this could have happened, but it had to happen. This, by the way, is why I get very impatient with all the moralistic finger-wagging, often enough by ferociously rich investment bankers, about profligate spenders living beyond their means and hard workers squirreling away their savings. This is mostly nonsense. If the US runs a trade deficit, for whatever reason, US capital imports must rise, and almost certainly that means debt will rise, and either it is producer debt or it is consumer debt. We don’t need hand wringing about declining morals to explain this.
By the same token, if China forcibly exports capital to the US, its savings rate must rise, not because Chinese households are being increasingly thrifty – in fact while China’s savings rate has surged in the past decade, the household savings rate has not, it was government and corporate savings that surged. I would in fact argue that if you include the wealth effect of negative real deposit rates, none of the increase in Chinese savings can be ascribed to household thrift but rather to policies that repress consumption.
So if Krugman is right, and the PBoC currency intervention is forcing China into a large capital deficit position, the most plausible consequences for the US must be either an equivalent rise in government investment or a decline in US savings. And of course it goes without saying that it must also result in a US current account deficit.
This is why Krugman has argued for a revaluation of the RMB. He is really just saying that the PBoC needs to do something that will reduce China’s massive capital account deficit, which will automatically bring down its current account surplus. Note that bringing down the capital account deficit is not necessarily the same thing as lowering the pace of reserve accumulation. Many people have warned that if the RMB begins to appreciate, it would set off hot money inflows that would force the PBoC into even more rapid reserve accumulation. While this is certainly true and certainly a major problem for the PBoC, what matters is China’s net capital account deficit. This consists of reserve accumulation less net capital inflows on the non-PBoC account.
Where is Krugman wrong?
So where could the flaws in Krugman’s argument be? Here are some areas we would need to consider:
- Perhaps the PBoC intervention is not driven by Chinese policies. Perhaps it is a residual of US policies that force the US into a capital account surplus position – because of a surge in US investment (which we know didn’t happen) or an endogenously-caused collapse in US savings. In this case the root cause of the imbalances in both countries is the collapse in US savings, and for reasons that are not altogether clear the PBoC has decided to accommodate this collapse in US savings by purchasing huge amounts of USG bonds, in spite of the distortions it introduces to the Chinese economy.
- Perhaps Krugman’s analysis is correct with regards to China and the rest of the world, but the US is not the rest of the world, and so while a Chinese revaluation would have the impact Krugman claims for the rest of the world in the aggregate, it would not necessarily have it for the US. In that case we would need to think about how different financial systems and industrial policies accommodate or refuse to accommodate capital account surpluses. My instinct is that open countries with very flexible financial systems are all likely to react in similar ways.
- Perhaps the Chinese current account surplus is driven by “natural” factors that cannot be changed and that have nothing to do with PBoC intervention. In that case PBoC intervention is simply a residual. One commonly heard argument is that the Chinese “naturally” save a lot, and so they must always run surpluses because savings will always exceed investment. At this point of the argument wholly mistaken references are often made to China’s late Ming and early Qing trade surplus with Europe, or someone comes up with a totally irrelevant story of how he met a Chinese family that is saving for their medical bills, son’s college, or so on. This argument I think is nonsensical, for reasons I explore above, and anyway it would imply that there would have been no need for the PBoC to intervene in the currency markets because intervention would have no effect on China’s trade surplus anyway, in which case why not just let the RMB appreciate? I always find this argument a little bizarre.
- Another common argument for natural advantage is that China has a huge surplus of cheap labor and so it must run a current account surplus. This is also sort of nonsensical since, aside from the fact that other countries with equally cheap labor don’t automatically run current account surpluses like China’s, the fact that Chinese labor is relatively cheap is also a function of domestic currency and wage policies, and anyway Chinese growth is capital intensive, not labor intensive.
- Perhaps however there is another, more plausible, natural advantage that causes the trade surplus. One candidate might be demographic transformations taking place in China, in the US, or both.
Stephen Roach’s counterblast
One of my favorite writers on global economics, Stephen Roach, in an article in the Financial Times had several rejoinders to Paul Krugman, some plausible, some less so. First off he argued against a bilateral approach to the trade issue.
Unless the problems that have given rise to the multilateral trade deficit are addressed, bilateral intervention would simply shift the Chinese portion of America’s international imbalance to someone else. That “someone” would most likely be a higher-cost producer – in effect, squeezing the purchasing power of hard-pressed US consumers.
Roach is right that this is more than a bilateral issue, and Krugman clearly agrees, but he is almost certainly wrong in assuming that bilateral intervention would simply shift Chinese exports elsewhere, nor am I convinced that it makes sense to call for an increase in savings while deploring anything that squeezes the purchasing power of consumers. I discussed both of these issues in one of my earlier posts.
Still, what is not clear, as he points out, is what the aggregate impact on the US trade account would be. It is easy to posit a number of plausible scenarios in which a significant contraction in the Chinese trade surplus might “only” show up as an expansion in the trade surpluses of Mexico, Vietnam or some other third country (although of course that country wouldn’t necessarily see this as a bad thing).
The US would be far better served if it faced up to why it is confronted with a massive multilateral trade deficit. America’s core economic problem is saving, not China. In 2009, the broadest measure of domestic US saving – the net national saving rate – fell to a record low of -2.5 per cent of national income. That means America must import surplus saving from abroad to fund its future growth – and run current account and trade deficits to attract the foreign capital. Thus, for a savings-short economy, there is no escaping large multilateral trade imbalances.
Yes, China is the biggest piece of America’s multilateral trade deficit. But that is because high-cost US companies are turning to China as a low-cost offshore efficiency solution. It also reflects the preferences of US consumers for low-cost and increasingly high-quality goods made in China. In other words, savings-short America is actually quite fortunate to have China as a large trading partner.
The last sentence is a little astonishing to me – the equivalent of Hillary Clinton begging China to keep buying USG bonds. This is basically the same thing as saying that the US is fortunate to have a large current account deficit, and begging that it be increased. But let’s ignore that. As I discuss above, to say the problem is too little savings in the US is to say nothing – it is true by definition and just part of the accounting identity.
It may very well be that American savings rates have declined for purely endogenous reasons, and that China’s capital exports are a residual impact, but at least for me it is easier to explain the fall in US savings at least partly as a consequence of the automatic adjustment necessary if there is an increase in Chinese savings. In the very next paragraph Roach makes just this point. He argues that China must reduce its surplus savings, as if these issues – low American savings and high Chinese savings – were independent. He makes the same point in his conclusion, “America needs deficit reduction and an increase in personal saving, while China needs to stimulate internal private consumption.” These are not separate issues. One can only occur with the other, so we are left with the problem of where the original distortion lies and how to resolve it.
Further on Roach says:
Yet some of America’s most prominent economists are claiming that a revaluation of the renminbi vis-à-vis the dollar would not only create more than 1m jobs in the US but that it would inject new vigour into an otherwise anaemic global recovery. Economists should know better. Changes in relative prices are the ultimate zero-sum game – they re-slice the pie rather than expand or shrink it.
He is of course right here. Ignoring the long-term impacts on growth in China and abroad, the currency game is a zero-sum game in the short term. It is a tug of war over employment, and that is exactly why China wants to maintain an undervalued currency and the US and Europe want it to revalue. This also suggests why, for all the recent signs of thawing, this issue is simply not going to go away as long as global unemployment is a problem.
I think probably the most important takeaway from Roach’s article is that this is a zero-sum game. We can resolve this problem concertedly and intelligently with the minimum cost to the global recovery (which means that Germany, Japan, and Europe have to be involved in the adjustment), or we can do so in a series of beggar-thy-neighbor confrontations. It is unlikely that the latter will involve the least cost to the global economy since the whole point of the strategy is not to minimize cost but to push as much of it as possible onto your neighbor..
And now, for something completely different
Before finishing, and in a bid to be topical, a quick word about the success of index futures, which were recently introduced into the Chinese stock market. Here is what an article in today’s Financial Times says:
Chinese investors love new financial products. Their passion, at times bordering on mania, explains why Shanghai stocks rocket in price on their trading debuts and why local equity funds can raise billions of renminbi on the day they launch. But even seasoned market professionals have been surprised by the enthusiasm with which investors have embraced stock index futures – China’s first financial futures since the mid-90s – following their launch two weeks ago.
“The volumes have exceeded everyone’s expectations,” says Dean Owen, Shanghai-based chief representative for Newedge, the French futures brokerage, which has a joint venture with Citic Group in China. Indeed, on Tuesday last week, the third day of trading, the value of stock index futures traded on the China Financial Futures Exchange exceeded the value of stocks traded on the Shanghai Stock Exchange.
This is part of an old discussion. Six years ago when QFIIs were first introduced into China there was a great deal of excitement about how the introduction of sophisticated foreign investors would help change the Chinese markets from being very speculative to being more sophisticated and value oriented. In a conference at which I spoke a senior official from the Shanghai stock exchange made exactly this point.
I disagreed with him. It seemed to me that the speculative nature of Chinese markets has nothing to do with the Chinese “love of gambling,” but rather is caused by the lack of tools available for value investing – macro data is questionable, financial statements are very poor, the corporate governance framework is at best mysterious, and the regulatory and policy framework is constantly shifting, often to achieve government objectives. Even Warren Buffet would give up trying to invest for value if he moved to China and traded A-shares.
On the other hand the market is extremely conducive to speculative activity. Speculators trade on short term changes in supply and demand factors, and in China stocks move rapidly for a number of non-fundamental reasons – changes in liquidity, regulatory and policy changes, insider activity, policy signaling, and so on.
At the conference, we agreed to disagree. I said in five years the Chinese stock markets would be as speculative as ever. He argued that it was already becoming more fundamentally driven and would advance significantly over the next five years. I am pretty sure nothing important has changed. The market is, if anything, even more speculative than it used to be.
Until the conditions that penalize fundamental investing and encourage speculation change, the Chinese stock market will be purely speculative no matter how many “sophisticated” investors or derivative instruments are available. A lot of people hoped that the introduction of index trading would allow investors to hedge and so somehow because of that would make the markets more fundamentally driven and stable.
This won’t happen. It is not because Shanghai lacks the accouterments of the NYSE or LSE that it is an unsophisticated and “speculative” market. It is because the tools value investors need – reliable information, clear corporate governance, a stable regulatory and policy framework, limited government interference – don’t exist. Index futures change none of that.
Financial liberalization and reform in China is only meaningful if it accomplished the following:
- Liberalize the setting of interest rates
- Clarify corporate governance at banks and large corporations – which means essentially make them subordinate to stockholders
- Improve macro and financial statement data
- Limit policy-driven government signaling, government interference, and regulatory changes
Without these, it seems to me that most reform and liberalization has been cosmetic.