Defending the RMB

It’s hard to short China, but not so hard to short China’s currency, and that’s a problem for the central bank.

We’ve all heard about PBOC intervention in the spot exchange market, where the central bank is selling some of its vast horde of USD Treasury securities and buying RMB (thus shrinking its own balance sheet, since RMB are its own liability).  A recent report suggests that the PBOC is also fighting back by trading onshore FX swap contracts, which the report characterizes as “unusual and complex financial derivative instruments”.

Actually, not so unusual or complex.  Let’s understand what this is all about.  (The next few paragraphs walk through some of the mechanics of these contracts.  People who know this already are invited to skip down.)

To start with, the Chicago Mercantile Exchange quotes a range of RMB/USD futures contracts, and provides extensive documentation of these contracts.  A recent settlement price for Sept 2016 futures was .15018 dollars/RMB, which is 6.6587 RMB/USD, at a time when spot RMB/USD was 6.4075.   That’s a difference of .2512, a sign that the “market” is betting the RMB will depreciate over the next year.

It is important to appreciate that, like other derivative contracts, futures contracts are zero sum bets.  If I short the RMB at the current futures price, then you or someone else must be taking the long side, and if the RMB depreciates by less than 2512 then I’m going to wind up losing, and my losses will be your gains.  In fact, every day from now until next September the futures price will move to equate current supply and demand pressures, so that every day one of us will win and one of us will lose.   A key feature of futures is that this daily gain or loss is settled daily, and if you can’t settle then your position is closed.

So much for the mechanics, now comes the economics.

It is vitally important NOT to fall into the trap of interpreting the futures price as an unbiased estimate of the likely future spot exchange rate.   That’s the theory of so-called Uncovered Interest Parity, and that theory is regularly and soundly rejected by the data.  Typical empirical studies find that, not only is the futures price a biased estimate of future spot, but more often than not it even gets the direction wrong!  If the futures price suggests the currency will depreciate, then it actually appreciates instead.

To be sure, speculators no doubt have some expectation about the future RMB/USD spot exchange rate.  But there is no reason to think that the futures price distills those individual expectations into a composite reflecting the “wisdom of the crowd”.  In general the (unobservable) market expectation of future spot price differs from the current (observable) futures price by a risk premium, which is largely a liquidity premium.  The market clearing price equates buying pressure and selling pressure, that’s all.

So now let’s talk about those FX swaps.

FX swaps work just like futures, except that they are not repriced every day, so that gains or losses are settled only at specific settlement dates.  At inception, the counterparties commit to trade RMB for dollars at a specific date in the future using a specified forward exchange rate, the forward rate, that is generally different from the current spot rate.  (So-called Covered Interest Parity is a typical reference point for determining that forward rate at inception.)

In the case of the RMB, there are two relevant FX swap markets, one for onshore (so-called CNY) and one for offshore (so-called CNH).  Apparently the PBOC has been intervening in the onshore market.  Here is the key passage from the report:

Thanks to what is described as “massive” orders from a few commercial banks acting on the PBOC’s behalf, the one-year dollar-yuan swap plunged to 1200 points from 1730 points Wednesday.  In the offshore market, the one-year dollar-yuan swap also dropped to 1950 points from 2310 points Tuesday, following the onshore swaps move.”

We can think of the Tuesday offshore dollar-yuan swap price of 2310 as somewhat analogous to the Sept 2016 CME futures price of 2512 that we talked about above.  We wouldn’t expect the exact same price because of the differences between futures and forward contracts, but it is pretty close.  The onshore price of 1730 suggests that selling pressure was less onshore than offshore, but in both cases selling pressure was stronger than buying pressure.

In this situation, what the PBOC apparently did was to enter as a buyer, but in the onshore market only.  That’s why the onshore price went to 1200–the forward rate is still weaker than the spot rate, but not by as much.

Arbitrage did the rest.

We do not expect the onshore and offshore markets to track exactly, if only because they reference different exchange rates (CNY vs. CNH).  But the markets are linked, and some people can trade in both markets.  That is why PBOC intervention in the onshore market got felt also in offshore, a move of 360 points rather than 530 points, but nevertheless a move.  And we can expect it to be felt also in CME futures, although again perhaps somewhat less.

Taking a longer view, remember that the PBOC held the CNY exchange peg with the dollar for a long time, even as the dollar appreciated against everyone else, and even as the Chinese stock market dropped precipitously.  But the August 11 decision to devalue opened the door to speculators who wanted to short China by shorting the RMB.   That’s what was driving the FX swap prices as of Tuesday.  On Wednesday, PBOC intervention moved those prices, and in doing so caused losses for those who were short, and gains for those who were long.

It’s a shot across the bow.  The PBOC is putting speculators on notice.  What happens next depends on whether speculators move on, or double down.   Unlike other major central banks, the PBOC has no swap line with the Fed.  But it does have a mighty war chest of dollar reserves, and is apparently willing to use it.  Betting against the RMB is not a sure thing.

4 comments on “Defending the RMB

  1. I’m a Coursera auditor and a fan. Didn’t comprehend everything in this post, but I am working at it!

    Best

  2. Homunculus on said:

    Loyal follower of the money view. Fed watchers need to look past Jackson Hole to China President Xi’s state visit to the U.S. We also ought simultaneously consider Kissinger’s view the U.S. and China lead in cooperation. Currency and markets are clearly front and center in this meeting. CBOCs latest statement on punishing profiteers indicates market making beyond its control. Both countries have a strong interest in more deeply regulating the off track betting equivalent of money which presents persistent risk in both economies. Let’s take another step towards regulatory stability.

  3. I posted this at Naked Capitalism but thought it should be here too.

    Sorry Professor Merhling, an FX swap is not the same as a future. A future is much closer to an outright forward, a very different FX beast.

    From Wikipedea:

    “In finance, a foreign exchange swap, forex swap, or FX swap is a simultaneous purchase and sale of identical amounts of one currency for another with two different value dates (normally spot to forward).[1] see Foreign exchange derivative. Foreign Exchange Swap allows sums of a certain currency to be used to fund charges designated in another currency without acquiring foreign exchange risk. It permits companies that have funds in different currencies to manage them efficiently.[2] swap contract: swap contract is an agreement between two parties to exchange a cash flow in one currency against a cash flow in another currency according to predetermined terms and conditions.”

    An FX swap is essentially a money market roll transaction with a buy spot, sell forward, or vice versa, depending on the intent. For a client: sell CNY spot, buy USD, one year forward. The Bank would then book: buy CNY spot, sell USD one year forward. FX delta risk here is minimal except in huge trades with big differentials.

    An outright is a future dated trade, also adjusted for the interest rate differential, but without any contra spot offset between the parties to keep it FX risk neutral. For the CNY one year today is 6.5140 with 1344 forward points on spot of 6.3778.

    The future has opposite trades in place, and it is zero sum, but both parties have FX risk and reward which is not the case in an FX swap where the parties have interest rate risk only.

    So if the PBOC had intervened in the swap market all it was changing was the market pricing for the difference between US and Chinese interest rates. The blowout in forward rates is an indication of USD funding stress in interbank, not spot market pricing of the yuan.

    • Perry Mehrling on said:

      I did not mean to leave impression that FX swap was the same as a futures contract, but I can see that informal language could leave that impression.

      There are really two issues in the difference between futures and FX swap, and I focused on only one of them. One issue is the difference between a forward and a future, and that’s the main thing I was talking about. The other is the difference between a forward and a swap, which is the main thing you are talking about. The swap does have a forward embedded, which is my emphasis, but it also has a simultaneous spot transaction going the other way, as you say.

      Your emphasis on the interest differential is, I think, equivalent to what I was saying with the forward-spot language. The link is covered interest parity:

      F = S(1+r)/(1+r*)

      where r and r* are respectively the dollar and foreign interest rate for a given term T, s is the spot exchange rate and f is the forward exchange rate for date T.

      The point of my post was that pushing around the forward price (in the forward leg of onshore FX swaps) winds up pushing around the offshore FX swap and also offshore futures price and so causing losses to shorts. I think that holds up.