August 22nd, 2015
in Op Ed
by Michael Pettis, China Financial Markets
Last Tuesday the PBoC surprised the markets with a partial deregulation of the currency regime, prompting a great deal of discussion and debate about the value of the RMB. Part of the discussion was informed by a consensus developing in one part of the market that the RMB is no longer undervalued but is in fact overvalued. Why? Because if left to the "market", that is if the PBoC stopped intervening, the excess of dollar demand over supply would force the RMB to fall.
This argument is based on a pretty confused understanding of how markets work and why investors do what they do. I thought it might be useful if I were to try to lay out the issue a little more clearly, and along the way address related issues. Because it isn't necessarily easy to tie all of the topics together in an essay, I thought it might be better if I put it in the form of a series of questions.
There are two conclusions, or at least two points I would argue:
- "Market" forces, that is the balance of supply and demand, do not always indicate the relative valuation of an asset. This is partly because there are several ways to define market forces, but mostly because we usually think of valuation in terms of economic fundamentals. An overvalued currency is one in which market fundamentals, by which I mean the valuation of assets on the basis of expected cashflows discounted at an interest rate that is not distorted, drive supply and demand.
- Supply and demand for an asset can also be driven by what traders often call "technical" factors. These are generally changes in supply in demand caused by other than fundamental factors. When China first approved the QFII program that permitted foreign investors to buy stocks, for example, or had China's stock markets been included in the MSCI global benchmark in June, as was expected, there was or would have been an immediate increase in demand for Chinese stocks that would have caused prices to rise for reasons that had nothing to do with an improved economic outlook.
When I was a student, I was taught that if prices did rise, they would do so by an imperceptible amount because they had been trading at a level consistent with a fundamental balance between demand and supply, and as soon as foreign purchasing caused prices to increase, Chinese investors would take advantage of "excessively high" prices to sell out. Of course this is almost the opposite of what happened. Prices rose precisely because of expected buying, and then fell in the case where the buying materialized.
There was no fundamental valuation to anchor prices. Once I became a trader this was one of the many things I had to unlearn, but rather than reject altogether the idea that fundamental valuation plays any role, which too often is the reaction traders have when they first learn that markets are not always driven by value, I thought it would be more useful to identify the conditions under which market prices do or do not respond to fundamentals.
- The RMB almost certainly would decline in value today without PBoC intervention, but this does not indicate at all that the RMB is overvalued. In fact the best argument is that the market is driven largely by technical, and that if we try to extract information from fundamental markets, we almost certainly would arrive at a very different conclusion. The RMB, it turns out, remains undervalued, although I suspect not by very much.
How did the PBoC change its currency regime?
The PBoC's statement on August 11 that it was changing the country's currency regime set off an explosion of analysis, accusation, praise and questioning that hasn't yet subsided much. Along with devaluing the currency by 1.86%, the most since 1994, the PBoC announced that it would modify the way it set the reference rate, known as "central parity", that determines the RMB's trading band, and it would so so
"for the purpose of enhancing the market-orientation and benchmark status of central parity".
It has, in effect, partially deregulated the exchange rate mechanism by relaxing intervention procedures, although it is still able to intervene as much as ever. Effective August 11, the PBoC said, the central parity would be set on a daily basis equal to
"the closing rate of the inter-bank foreign exchange market on the previous day."
Probably to indicate that this did not mean the end of PBoC intervention, it added that the rate would be set
"in conjunction with demand and supply condition in the foreign exchange market and exchange rate movement of the major currencies."
Until that day the PBoC set central parity every day at whatever rate it thought appropriate. In principle this is supposed to mean that the value of the currency is a function of the PBoC's best estimate of the exchange rate that maximizes China's long-term productivity. In the best of cases, however, the sheer complexity of any economy, let alone the global economy within which it operates, would make this impossibly difficult to determine even if there were objective ways of valuing the choice between a short-term cost or benefit and a long-term cost or benefit, or of choosing how costs or benefits will be distributed among different economic sectors or social groups.
This is why there is a grudging consensus, although certainly not unanimous (nor is all the consensus grudging), that the most effective and efficient way to determine the exchange rate is to let the market decide. If all potential buyers and all sellers of RMB, whatever their reasons, collectively decide on a price at which all transactions can clear, that price is presumably the best estimate of the exchange rate that maximizes China's long-term productivity.
Why did they do it?
There are three different reasons that might explain the PBoC's move Tuesday.
- Improve trade. While China's current account surplus has been very high, this is mainly because imports have done worse than exports. Both have fared poorly. The numbers were especially bad in July, when imports were down 8.1% year on year while exports were down 8.3%. Because of its peg to the appreciating dollar, the renminbi has been very strong on a trade-weighted basis. The new currency regime may be aimed at reversing this.
- Qualify for SDR. There may have been concern that the large and persistent gap between the fix and actual trading in both the onshore and the offshore markets would prevent the RMB from qualifying for inclusion in the SDR basket. What is more, by including a RMB pegged to the dollar, the already overly dominant weight of the dollar in the SDR would be substantially increased, something the IMF clearly does not want. Beijing may be eager for the RMB to become part of the SDR basket because it believes this will result in significant foreign inflows that will help reverse China's very large and potentially destabilizing capital account deficit. Its strategy may be working. On Wednesday the IMF described the new pricing mechanism as "a welcome step as it should allow market forces to have a greater role in determining the exchange rate". It followed by noting, a little obviously, that the "exact impact will depend on how the new mechanism is implemented in practice".
- Monetary freedom. The well-known "impossible trinity" makes it impossible for a central bank to control both domestic interest rates and the exchange rate if its capital account is open. Although technically not open, China's capital account is porous enough for all practical purposes. This means that as long as the PBoC intervenes in the currency, it cannot provide debt relief to struggling borrowers, and to the economy overall, by lowering interest rates without setting off potentially destabilizing capital outflows. This constraint would be even tighter if the Fed began to raise interest rates. Reform of the exchange rate mechanism restores interest rate flexibility.
There is no way to say for sure which of these drove the PBoC decision because the PBoC, like most central banks, has preferred to be a little vague about its reasoning, but I suspect that it changed the currency regime primarily either to gain monetary freedom or, more likely, to qualify for inclusion in the SDR. I doubt that the desire to turbocharge exports played much of a role, but I worry that if the PBoC was hoping to reverse the huge deficit on its capital account, the success of its plan will hinge on whether it was able to distinguish between fundamental demand and technical demand.
What determines the PBoC's best estimate of the appropriate exchange rate?
The exchange rate has several functions in any economy, and what the PBoC decides is the most appropriate exchange rate depends on what it is trying to accomplish. These include:
- By transferring wealth from one sector to another, the exchange rate can be used to subsidize favored sectors at the expense of others. High exchange rates benefit household consumers, the services sector, and urban residents, among others. Low exchange rates benefit the tradable goods sector and commodity producers, among others.
- This process of transferring wealth also means that a higher exchange will speed up the rebalancing process, in this case by transferring wealth from the PBoC and the tradable goods sector to households.
- The interest rate and the exchange rate are two of the most important prices in an economy, or put differently, they are two of the most important pieces of information economic entities, including businesses, use to make investment and operating decisions. When the RMB trades at its fundamental value, economic agents within China are most likely to make the decisions that optimize overall value today and preserve the sustainability of economic behavior. Anything that pushes it away from fundamental valuations will distort the investment and operational behavior of all economic entities.
How does the market provide information?
We can usefully think of the "market" as a machine that processes a vast amount of information quickly and smoothly. Any agent who possesses superior information about a product or service that will cause a change in its supply or its demand will buy or sell based on the information. His buying or selling becomes the way in which the information is absorbed by the market and presented, in the form of a price, to all other agents.
This isn't necessarily the most accurate of ways in which to determine a price but it seems to be more accurate than any other method we have been able to come up with. Put differently, it seems to be the most accurate way to drive the allocation of goods and services in a way that maximizes social wealth (which is, after all, what a market is supposed to do). We shouldn't assume, however, that with the market all of the tough questions have been adequately answered.
The market implicitly does determine the answer to these questions, like the trade-off between the present and the future, or the different values it places on the needs of different social groups. We know that the market does these things because that is what it means for the market to clear, but the answers it provides will vary according to the institutions, including moral values, that form part of the system within which it operates.
One of the most important advantages, or efficiencies, of "letting the market decide", however, is that it doesn't seem like the answers the market gives us are in fact affected by our institutional setup. The market's "decisions" are given a veneer of neutrality that everyone accepts, even though the decision is not neutral at all. This seeming neutrality reduces the political maneuvering that might otherwise occur.
Does PBoC intervention to support the RMB mean that it is overvalued?
One of the main questions being batted around is whether, under the new system, the value of the RMB is finally going to be determined by the market. If it is, it almost certainly means that the value of the RMB will decline.
Why? Because the balance of payments, which is the sum of the current account surplus and the capital account deficit, is in deficit if we exclude PBoC interventions. At current prices there is more RMB selling than there is buying, and the PBoC has to sell reserves and buy RMB in order to keep the currency from depreciating.
This, many people argue, proves that the RMB is overvalued. The "market", they claim, has spoken, and it has told us that the RMB is overvalued.
They are wrong. The "market" is not telling us that the RMB is overvalued. It is telling us only that there is more supply of RMB than there is demand for RMB at the current exchange rate. Because "overvaluation" and "undervaluation" usually refer to the fundamental value of a currency, this excess of supply over demand would only imply an overvaluation of the RMB if supply and demand were driven primarily by economic fundamentals.
What drives supply and demand for the RMB?
Excluding central bank intervention, which is mainly a residual contributed automatically by the PBoC to balance supply and demand for foreign currency, all purchases or sales of foreign currency in China can be divided into current account activity, which mostly consists of the trade account, along with other transactions including tourism, royalty payments, interest payments, etc., and capital account activity, which consists of direct investment, portfolio investment, and official flows.
Imbalances in both the current account and the capital account can be driven by economic fundamentals, in which case it might make sense to say that the RMB's "correct" exchange rate is broadly equal to the clearing price at which supply is equal to demand. In this case if the central bank were to purchase RMB, reserves would decline and it would be reasonable to assume that PBoC intervention would cause the RMB to become overvalued, while PBoC sales of RMB would cause reserves to increase and the RMB to become undervalued.
But neither the current account nor the capital account is necessarily driven only by economic fundamentals. As an aside, most people, including unfortunately most economists, typically assume that the current account is independent and the capital account, if they think of it at all, simply adjusts to maintain the balance, but this is an extremely confused way to think about the balance of payments.
During the past thirty years especially, reserve accumulation and private capital flows have overwhelmed trade flows, to the extent that small changes in gross capital flows have often forced large changes in trade and current flows. Even in the 1950s and 1960s, when international capital flows were much smaller, and did not drive trade flows to nearly the same extent, capital flows very often constrained trade flows - most famously the "dollar shortage", which was only relieved by the Marshall Plan.
What might push the current account balance away from fundamentals?
If the domestic and foreign tradable goods sectors are not subsidized or penalized, the market for tradable goods can be seen as operating largely on the basis of fundamentals. In that case imbalances in the supply and demand for foreign currency may indicate under- or over-valuation of the currency. But the trade account can depart from fundamentals under these or similar conditions:
- If Beijing were to impose significant tariffs, prohibitions on imports, or otherwise intervene directly in trade, it could distort the current account in ways that do not reflect economic fundamentals.
- Regulations that discriminate against or in favor of imports - for example Japan in the 1970s and 1980s imposed safety inspections for foreign cars, whose very slow pace prevented imported automobiles from being competitive except at the very high end of the market - can distort fundamentals.
- Without central bank intervention the capital account must balance the current account, so that if there were significant net inflows or outflows on the capital account, these net flows would force the RMB up or down in order that the current account surplus or deficit balances the capital account.
What might push the capital account balance away from fundamentals?
If capital enters or leaves China in order to earn higher expected returns on investment in business or manufacturing capacity, or to purchase undervalued assets, we can broadly assume that capital flows are driven by fundamentals. Otherwise they are not. Non-fundamental capital flows into or out of China might include:
- Wealth or portfolio diversification,
- Speculative inflows driven by asset bubbles, or "carry trades" driven by arbitrage opportunities based on capital-flow restrictions,
- Flight capital driven by fear of political instability, financial instability, or the anti-corruption drive,
- Official flows driven by political considerations,
- Investment outflows that are not sensitive to economic fundamentals, for example those driven by government directives to acquire commodity-producing businesses and land, or to acquire technology or management skills
Can the market decide on an appropriate value for the RMB?
Whether or not we believe that the market should determine the value of the RMB is one of those questions - like whether or not we support of free trade, supply side economics, fiscal deficit limits, debt, etc. - that tends to be framed as a question of principle when in fact it isn't. These work well to enhance productivity and wealth under certain conditions and fail under others, so that it is much more useful to specify the conditions under which they work - for example, what are the conditions in which China would generally be better off if the markets decided the value of the RMB?
To answer we need first to decide what our objective is. If we have political goals, for example wealth redistribution, or the protection of certain types of industries until they are sufficiently competitive, we would probably want to start with an idea of the economically optimal exchange rate on a fundamental basis and then move it in one direction or the other.
In China's case, I would argue that the goal is to eliminate some of the distortions in the Chinese economy that weaken domestic demand and systematically misprice economic inputs, most notoriously capital. These have left China with a dangerous dependence on debt, excess capacity and inventory, and a state sector in which incentives to innovate and create value are overwhelmed by political incentives (that include capturing explicit or implicit subsidies).
One consequence has been so much wasted investment that I am convinced that many years from now we will look back at China in the 2000s, rather than Japan of the 1980s, as the classic example of capital misallocation. If eliminating these distortions is indeed the goal, I would argue that the correct exchange rate would probably be one that is determined by the country's economic fundamentals, i.e. one that matches supply and demand for dollars in the real economy - or perhaps a little stronger than that in order to help the rebalancing process.
If on the other hand the goal is to ensure that China has sufficient reserves, I would argue that the correct exchange rate would probably be one that is determined by the country's overall balance of payments. In the past a country's money supply was often a function of its gold or silver reserves, and economic performance could be severely impaired by a shortfall of specie reserves. Today there are countries running persistent deficits, or in which domestic investment is severely constrained by insufficient savings. In these two cases it might make sense to focus on the overall balance of payments and the information it might give us about an appropriate exchange rate for the RMB, but China is obviously not one of these countries.
What happens if the exchange rate is set at a "wrong" price?
An important characteristic of a market is its systemic ability to adjust, whether quickly or not. If there is a distortion in the price of any good or service, the price of other goods and services automatically adjust to return the market to what is assumed to be an optimal stage.
This is why economists who argue that the value of currencies like the RMB should be fixed - usually in terms of other major currencies, such as the dollar, or in exchange for commodities, the longest serving of which has been cowries, followed by gold - can also argue that markets should determine all prices without being inconsistent. If the central bank pegs the value of its currency to another currency, as the PBoC pegs the value of the RMB to the USD, all other relevant variables, most importantly the interest rate, will automatically adjust so that the economy will presumably get the full benefit of the market's superior ability to process information.
How do prices adjust to distortions?
Every transaction or policy moves a system away from equilibrium, just like every price distortion. If there are reasons to prevent a quick return to equilibrium, the system becomes increasingly unbalanced. This is why Albert Hirschman argued that all growth tends to be unbalanced growth, and in economies with very large state sectors, these imbalances can persist.
In the idealized "Smithian" world of innumerable economic agents none of which is big enough to distort the adjustment process, the economy must quickly adjust so that the system is never far from equilibrium. In this world the only thing that can cause a crisis, which is a nothing more than a very rapid, disruptive adjustment of a major imbalance, is some kind of major external shock, which is soon followed by a crisis.
In a world, however, in which there are institutions or institutional players large and powerful enough to block the adjustment process, the imbalances can build for a very long time. As they do, they put increasing pressure on the institutions that prevent adjustment, and so the adjustment itself becomes increasingly disruptive (often causing policymakers to react by preventing adjustment even more aggressively, thus locking the country into a potentially self-reinforcing process of growing imbalances). Eventually the adjustment must occur, either rapidly in the form of a crisis (and it takes an increasingly small external "shock" to trigger the crisis), or slowly in the form of a long and usually difficult period of rebalancing.
This is not the place in which to enter into a long discussion of how economic systems work, but it should be clear that we live in a world in which there are many large institutions, most obviously governments, as well as legal and regulatory constraints, perhaps most importantly within the financial system, that prevent automatic adjustments from occurring immediately.
There are however at least two important points worth reminding anyone thinking about how currencies are pegged:
- Volatility is transformed, not eliminated. All economies are volatile, and while the impact of volatility on any economic entity can be exacerbated or minimized by the structure of its balance sheet, this only happens as volatility is assigned to agents less able or more able to absorb it. Pegging the RMB to the USD, for example, does not eliminate the volatility associated with expected changes in the USD value of the RMB. Instead the volatility shows up as higher volatility in China's money supply, higher volatility between USD and non-USD currencies, greater trade imbalances, and so.
- Interventions are effectively forms of wealth transfer. Pegging the RMB to the dollar at a low rate, for example, transfers wealth from importers to manufacturers in the tradable goods sector in a process well understood by most economists. But while it reduces currency volatility, it increases volatility in the money supply. What is more, as the PBoC attempts to control the interest-rate component of this volatility by fixing interest rates (which until recently also transferred wealth from savers to borrowers directly), there is even more volatility in China's money supply, both in the present, in the form of inflows and outflows, and in the future, as it is "stored" in the form of rising bad debt.
Because regulators can never choose how much volatility they will permit, at best they can choose the form of volatility they least prefer and try to control it by transferring it elsewhere. This is usually a political choice and not an economic one, and is about deciding which economic group will bear the cost of volatility.
Even when it is an economic choice aimed at resolving a particular problem, once that problem is resolved and the transfers begin to undermine productivity, the beneficiaries are often powerful enough to prevent reform. Government interventions in the currency usually aim at creating wealth transfers to subsidize favored sectors or at suppressing volatility that penalizes favored sectors, or both. The analysis of their impacts is never complete until we have also worked out the impact on those sectors from whom wealth has been transferred or to whom volatility has been transferred.
The "impossible trinity" adjustment
The best-known of these adjustment processes, and the one most relevant to the RMB, is the impossible trinity, which is simply a restatement of the way money supply must automatically adjust. In an open system (free capital flows being one of the three legs of the trinity) the PBoC can choose to peg the USD value of the RMB, in which case the money supply adjusts as money is created or destroyed in order to match supply and demand in the market in which RMB and USD are exchanged. Or it can chose to determine the size of the money supply (which it attempts to measure by looking at interest rates), in which case the exchange value of the RMB will rise or fall in order, once again, to match supply and demand in the market in which RMB and USD are exchanged.
This in fact may be one of the main reasons (which I listed above as the goal of regaining monetary freedom) the PBoC changed its currency regime. For the past two years Chinese interest rates have been too low relative to the value of the currency for supply and demand in the capital markets to balance, and we know this because China has a large capital account deficit. It experienced massive net outflows on the capital account.
Because these net outflows put destabilizing pressure on a banking system used to net inflows, there were two ways the PBoC could manage the process. It could raise interest rates high enough to satisfy investors, or it could cause them to reduce their required yields.
But it is important to understand that there is no particular reason in principle for supply and demand in the capital markets to balance. Before 2014 China ran large surpluses on both its current account and its capital account, and because the balance of payments must balance, by definition, it had elsewhere to run a massive deficit equal to the sum of the two, and it did in the form of a central bank deficit - another name for rising foreign exchange reserves.
The increase in central bank reserves was a residual, and not a decision by the PBoC. When it decides to peg the value of the currency, it has no choice but to accumulate or lose reserves, as the impossible trinity ensures that money supply rises or falls to match supply and demand in the market in which RMB and USD are exchanged.
The structure of the investor base matters
The fact that the capital account deficit has grown to overwhelm the current account surplus does not tell us whether or not the net imbalance is driven by fundamentals. What matters is whether or not the capital account is driven by fundamentals.
There are, very broadly, two reasons to bring money into China and two reasons to take it out, and we can define these as fundamental and speculative. As I explain in my book, Avoiding the Fall, as well as in a November 2013 blog entry, there are three different types of investment strategies that explain most investment decisions, and depending on the mix of investment strategies in any given market, the behavior of that market, including what it decides is information, determines whether that market will react to fundamental or technical information how it will interpret that information.
A fundamental investor brings money into China in order to invest in a project that will deliver value over the long term. A speculator brings money into China in order to purchase an asset, usually stocks or real estate, which he can quickly sell at a profit. Investors who borrow USD or HKD to buy short-term RMB-denominated government bonds or other debt in order to earn the interest rate spread, as well as profit from any increase in the value of the RMB, are technically relative value investors, but for our purposes are speculators because they provide net inflows into China if is seen separately from the offshore markets.
More importantly, they process and interpret information in the same was as speculators and rather than act to stabilize prices (by buying for value when prices decline), they tend to enhance volatility by reinforcing appreciation and depreciation expectations. Technically the second and third of the three are illegal and violate capital restrictions, but these involve domestic investors, businesses or SOEs who are able to take advantage of corruption or weak regulation to circumvent these restrictions.
Similarly a fundamental investor takes money out of China in order to invest in foreign projects that will deliver value, including diversification benefits, over the long term. Investors who take money out of China in order to hide it, however, or because they are frightened by perceived financial or political risk, should for our purposes be classified as speculators, not because they seek speculative profits but because they have the same systematic impact as speculators.
Finally there are investors who take money out of China in order to achieve political objectives. For example they may seek to reduce China's dependence on foreign-owned agricultural or non-agricultural commodities.
The purpose of this classification is not to identify the good guys and the bad guys but rather to understand market dynamics. I spent most of my career on Wall Street trading floors, and like most traders and institutional investors I think of markets differently than do most economists and policymakers. In order to understand how markets will perform I try to work out the structure of the investor base, understand market "technicals", i.e. potential changes in supply and demand, along with their triggers, and look for convexities or implied options. A market dominated by speculators must react in a profoundly but predictably different way than one dominated by fundamental investors.
How do speculators interpret information differently from fundamental investors?
One of the reasons the PBoC may have permitted RMB depreciation is to regain control of monetary policy. The "impossible trinity" prevents a central bank from controlling both domestic interest rates and the exchange rate if its capital account is open. Although technically not open, China's capital account is porous enough for it to be "open" for all practical purposes.
It turns out that interest rates in China are higher than they are elsewhere in part because of the constraints imposed by the impossible trinity. Even with higher interest rates on its government bonds, in which the risk of default is widely perceived to be close to zero, there is nonetheless a large net outflow on China's capital account. Why don't more investors take advantage of higher Chinese interest rates and equally low credit risk by bringing money into the country?
The most obvious reason is that they are worried about depreciation risk. Because most wealthy Chinese seem to think about RMB in terms of USD or Hong Kong dollars, it is the fear that any depreciation of the RMB against those two currencies (the Hong Kong dollar is pegged to the USD through a modified currency board) greater than the couple of percentage points interest rate differential would yield less than equivalent USD or Hong Kong dollar bonds.
This means that as long as the PBoC intervenes in the currency, it cannot provide debt relief to struggling borrowers, and to the economy overall, by lowering interest rates without setting off potentially destabilizing capital outflows as the interest rate differential narrows. This constraint would be even tighter if the Fed began to raise interest rates, which would also cause the interest rate differential to narrow.
So how do we reconcile the PBoC's desire to reduce interest rates with the higher interest rates investors need to compensate for the greater risk of devaluation? The answer, it turns out, is fairly straightforward. The interest rate investors require to buy bonds must decline until it is equal to the PBoC's target interest rate. Because the interest rate investors require is a function of their perception of the devaluation risk, this means that the currency must decline until the perception of devaluation risk is low enough to meet the PBoC targeted interest rate.
In that case it would be a fairly easy matter to reduce the value of the RMB to the point at which investors believe the currency to be correctly valued. Once it reaches that level, there is no longer a bias to currency volatility and the RMB is as likely to rise as it is to decline. The currency would then be fairly priced, the expected volatility very low and unbiased, and investors would require nothing more than the risk-free cost of capital (assuming, of course, that expected inflation is positive).
But is the capital account buying dollars for fundamental reasons - that is, because foreign assets are cheaper that Chinese assets or foreign growth expectations higher than Chinese growth expectations? Probably not. Three things seem to drive the outflow, which was a net inflow two years ago and has only recently surged to such astonishingly levels.
- Government-directed purchases of commodity producing assets or of strategic technologies, which are not sensitive to issues of fundamental valuation.
- Flight capital, driven probably by rising political or financial uncertainty, which is not sensitive to issues of fundamental valuation.
- Speculative capital worried about currency depreciation.
The capital account does not seem to be driven by fundamentals, in other words, and is instead driven mainly by outflows that are not sensitive to valuation issues. But in a highly speculative market, price movements are usually self-reinforcing, so that a falling RMB may actually increase the desire or need to sell. This might be because there are leveraged investors, including investors in derivatives, whose borrowing costs are inversely indexed to the exchange rate. More likely, it may also be because speculators interpret a dramatically falling RMB as signaling a change in PBoC policy and higher risk, so that the more the RMB depreciates, the higher the required premium. Finally, in a speculative market, if investors believe that they are collectively big enough to set off a self-reinforcing selling process, they may self-consciously decide to interpret a declining RMB as a sell signal. This last, especially, is well understood by traders, even when non-traders ssometimes find it too "irrational" to be credible.
In theory this means the value of the RMB could fall infinitely, but in practice it can only fall until it is low enough to bring out enough fundamental investors to convert the market from a speculative market to a fundamental one. Their buying then stabilizes the market and can actually set off a self-reinforcing price reversal. Alternatively the decline might be halted by very sophisticated interventions by the PBoC that convince speculators that the currency is more likely to rise, and so have the same impact.
So is the RMB overvalued or not?
Capital is leaving China for reasons that have little to do with economic fundamentals and that do not imply that the RMB is overvalued, and the capital account deficit is large enough to overwhelm the current account surplus. This suggests that the balance of payments is unbalanced, before PBoC intervention is factored in, but this imbalance tells us little about the fundamental value of the RMB.
On the other hand the fact that the trade account is in such large surplus seems to tell us that the RMB is undervalued. Why? Because if China is growing much faster than its trading partners, and if it has much lower unemployment than its trading partners, and if it is leveraging up while its trading partners are deleveraging, standard trade theory tells us very clearly that absent intervention and distortions, China would run a trade deficit, probably even a large one, and its partners the corresponding surplus.
But China is, instead, running a trade surplus. This "surprising" trade position should be a very clear indication that either the currency is undervalued, or that there is some other equivalent trade distortion. In itself, it seems fairly clear, at least to me, that the current account surplus indicates that the RMB is undervalued on a fundamental basis, and that the balance of payments deficit is caused primarily by speculative outflows, or other kinds of outflows that are not sensitive to economic valuation issues.
How does the RMB affect inflation?
One final point concerns the impact of China's devaluation on global deflation. As of this writing the RMB has depreciated by too small an amount to matter much, but assuming that it had depreciated by a lot more, would this add to global deflationary pressures?
Most analysts say that it would. A depreciating RMB causes the price of Chinese goods to fall, and so lower prices add to deflationary pressures. While I think it would indeed add to deflation, this is not because it reduces the price of Chinese goods. Ultimately deflation requires that aggregate supply for goods and services rise relative to aggregate demand, or that aggregate demand fall relative to aggregate supply.
If Chinese prices drop, how does it affect supply and demand abroad? By reducing the price of Chinese imports, it represses the tradable goods sector, which may respond by firing workers. It also raises the real value of disposable household income and in so doing may increase household consumption. The net impact depends on whether or not the consequent increase in the household income share of GDP is overwhelmed by the increase in unemployment.
How does it affect supply and demand in China? By increasing the price of foreign imports, it subsidizes the tradable goods sector, which because unemployment in China is low, may respond by bidding up wages. It also reduces the real value of disposable household income, which can reduce the consumption share of GDP. The net impact depends on whether or not the consequent increase in the household income share of GDP is overwhelmed by the increase in unemployment.
By itself a depreciating RMB is not deflationary for the world. It is only deflationary if it causes a relative increase in supply over demand, and this is most likely to occur because of the impact of consequent wealth transfers.
Academics, journalists, and government and NGO officials who want to subscribe to my newsletter, which sometimes includes portions of this blog and sometimes (as in this case) does not, should write to me at firstname.lastname@example.org, stating your affiliation, please. Investors who want to buy a subscription should write to me, also at that address.