The AUD has risen against USD (and in fact against developed currencies in general) in the global risk rally of the past two months, having previously dropped on sharp interest-rate cuts by the RBA. I wonder whether there could be a shorting opportunity on the way.

What moves AUD/USD? The strongest correlation seems to be with the S&P 500 — not rate differentials or copper (as a proxy for China) or anything else. I read this as meaning that AUD rises when a risk-on mood leads to yield-seeking capital outflows from commonly-used funding currencies.

What does this mean for a trade in AUD? It means that, if you short AUD, you are betting on a pullback in global risk appetite OR a temporary change in the drivers of AUD such as the market experienced in Mar-Apr 2012, when the RBA’s shift to a loosening bias led the AUD to fall vs. SPX and in absolute terms, OR both.

Over a shorter time horizon, my point about the correlation between AUD/USD and SPX holds, but over the longer-term, the interest-rate differential between AUD and USD does matter — which is why the RBA’s move had an effect. I think of the interest-rate differential as setting the broad level of AUD/USD and global risk appetite as determining its short-term gyrations. The market has recently gyrated upwards, such that a short position entered today would have a stop close to, but not within, the highest recent short-term trading band of AUD/USD (i.e., the extreme of the market’s range with interest rates at their previous level). Since interest rates are now significantly lower than they were when this range was made (in the first two months of 2012), it would be incoherent for the market to return to it unless the variance of the market has increased markedly for some other reason — and I see no reason to think that that has happened.

That last point establishes that there will be a reasonable entry point into a short position if the stop is in the range Jan-Feb 2012. But is does not establish that going into the market is a good idea. For that, one also has to have a reason why the trade should be expected to make money, and that takes us back to the two possible drivers of AUD that I identify above.

First, is it reasonable to bet on a pullback in global risk appetite? Logically, it should be. The markets were driven down partly by a worsening picture in Spain but mainly, I thought and still think, by the Greek situation and the attendant fear of a breakup of the Eurozone. This point of view is supported by the fact that SPX has recently returned to its trading range of Mar-Apr 2012, which was a new post-GFC high for the market, following the victory by pro-bailout parties in the Greek election. Greece and Eurozone breakup have now returned to the newsflow thanks to comments from the IMF about cutting Greece’s funding, a surprisingly hard line from Germany in the light of earlier intimations of a willingness to compromise with the government, and the Troika’s visit to Greece (in order to write its September progress report); further, other potential drivers of global risk including the US and Chinese economies look worse now than they did in April, and the US earnings season has been uninspiring and, from a revenue point of view, dire. It would not be coherent for markets to return to their Spring highs against this deteriorating backdrop.

On the other hand, central banks have acted. The ECB and BoE have taken different measures to deal with the shortage of high-quality collateral, with the BoE instituting a funding-for-lending programme and the ECB simply cutting its collateral standards again; and, on the monetary-policy front, the ECB has cut rates and the BoE has started another QE programme. I do not think that these measures alone will be enough to keep global risk appetite buoyant in the face of the negative factors that I mention above. But the Fed is also moving closer to another round of QE, and the ECB could do more.

Will the Fed act? To answer this question, we need to choose a model.

  • At the start of the year, Bernanke said that the unemployment and inflation targets were equally important, which both the market and I took to mean that the Fed would act if the employment situation ceased to improve as long as inflation was not dangerously above 2%. As I said at the time, it was not clear whether this form of the Bernanke put was a put on the rate of improvement of unemployment or a put with a strike that trailed the level of the unemployment rate, but the former idea can now be discounted because the rate of improvement slowed some time ago and the Fed has not acted. Is the latter a good model for the present situation? Unemployment ticked up a little in May and was flat in June, and it may be that May’s little tick was not enough to touch the trailing strike of the Bernanke put, so I do not think there is enough evidence to reject the model on the basis that the Fed has not acted. However, recent comments by Bernanke and other officials have suggested that the Fed is concerned about the costs of a further round of QE (about which they have been vague, but which include a further reduction of the pool of high-quality collateral). In other words, they seem unconvinced that the benefits for the employment situation would outweigh the costs of QE at present. Hence, I think that Bernanke’s comments in January are best interpreted as propaganda and that the unemployment-put model is not a good model for predicting what the Fed will do next week.
  • How else might we attempt to predict the Fed’s actions? An obvious thing to do is to look at how it has acted in the past few years in the context of its dual mandate. I have constructed a simple composite index using the 5-year TIPS breakeven rate and the monthly change in non-farm payrolls to represent the dual mandate. Let us have a simple model in which the Fed’s options range from “do nothing” through “other unconventional measures” to “QE”. The index is at around the same level as in mid 2011, when the Fed announced OT and introduced its mid-2013 language but declined to do more QE, and is well above the mid-2010 low that sparked off QE2. The Fed might anticipate some further deterioration in the index, but such deterioration would have to be very sharp to return the index to its 2010 lows before the September FOMC meeting. That suggests that we are in “other unconventional measures” territory.

In summary, the Fed’s communications have recently led Fed-watchers around in circles and left us with no sensible model of its current reaction function. It therefore makes sense to use a model of its historic reaction function, and my simple model based on a composite index suggests that further QE is not likely.

Are there further unconventional measures that the Fed could take? I am not aware of any that are attractive. OT, on current plans, will reach its limit in December because there will be no more short-term securities to sell; I suppose the Fed could twist further out on the curve, but that could lead to market distortions at the long end as the Fed comes to own a rising proportion of the longest-dated securities. Given that the effects of OT are probably pretty marginal, I suspect it is unlikely that the Fed will push it any further. Another alternative is for the Fed to extend its “expectation” of low rates beyond the end of 2014. That is possible, but the marginal effect of this change would be less than that of the first two “expectation” announcements as the yield curve has already flattened significantly, and its marginal credibility would be lower (the further out you go, the murkier the future gets and the less the “expectation” is relevant). Other potential measures, such as raising the inflation target, the specification of triggers for policy tightening or taking interest rates negative, are at present viewed with enough suspicion by FOMC members and politicians that their use is unlikely in the short term.

If the Fed’s options are limited when it comes to other unconventional measures, does that mean it will prefer to do nothing, or go for QE as an alternative? Given what Bernanke et al. have recently said about the costs of QE, I am inclined to think that they will avoid it unless there is a strong argument in favour. At present, the model does not make one. In conclusion, then, I think it is most likely that the Fed will hold policy at the coming meeting, and that there is some possibility of an extension of the “expectation” language beyond 2014.

What about the ECB? Mario Draghi said today that the ECB would do whatever was needed to preserve the euro — which is not a surprise, to me anyway. The ECB is an institution of European integration. He also said that “To the extent that… sovereign premia hamper the functioning of the monetary policy transmission channel, they come within” the ECB’s mandate. Well, that point is not new either, and sovereign premia were clearly hampering the transmission of monetary policy two months ago and the ECB did nothing about it. But central banks speak in code, and the ECB takes a particular delight in subtle hints. The markets have certainly taken the view that Draghi has indicated that further sovereign bond purchases by the SMP are likely soon. What will that mean for the markets? I tend to think: not too much. As I say above, I think that the markets’ decline and then rally were mainly about Greece, not Spain. The Spanish situation is not a pressing emergency — the country’s average funding cost remains around 4% — steps are being taken top deal with its banking problem, albeit slowly, and the market has lost interest. Greece is the problem, Greece has recently come back onto the radar screen, and Greece cannot be fixed with ECB bond purchases or LTRO’s. SMP purchases might allow the markets a short-term rally, but I don’t think they will be enough to create a trend.

It seems reasonable to think that global risk has rallied on the back of the Greek election and various European measures including the interest-rate cut, and that it is being supported by hopes of Fed action next week. I think that the Fed is likely to disappoint. Without some other source of support, risk assets may well fall if that disappointment comes to pass.

Second, is it reasonable to expect a temporary change in the drivers of AUD — i.e. further rate cuts by the RBA? That is not something I want to bet on explicitly. However, I think that rate hikes from the RBA are unlikey and that, with the Australian household sector cash-flow negative, the Chinese economy apparently continuing to slow and 3-year breakeven inflation only just into the target 2-3% range, further deterioration in the economy and further cuts in rates are a possibility. The best way to look at this is as a potential bonus: I don’t think a short AUD trade can stand on this basis alone.

In summary: markets are heading upwards and it may be possible to get a stop in the Spring highs of AUD/USD. But compared to the Spring, the economic situation is worse and the central bank that could really turn things around — the Fed — is unlikely to act. That means that markets ought not to rally much further, and that the risks are to the downside.

Risks to this view:

  • Harder rhetoric on Greece was actually all about getting Greek politicians to agree the specific 11bn of cuts — which they have now done.
  • ECB action does lead to a risk rally.
  • The Fed institutes QE3.