Tuesday, December 18, 2007

Reality bites

You know markets are feeling the squeeze when volatility is maintained in the week before Christmas. Usually this is a week to nip out shopping, enjoy long boozy lunches, and engage in quizzes with colleagues and counterparties. This year, however, there seems to be a fair amount of portfolio cleansing, with risk across a number of popular strategies getting cut. The upshot is that most directional players who made money last month are probably losing in December...as are many of those who lost money last month as well.

It's also the time for banks, strategists, and even bloggers to begin sending out their list of "Favourite trades for 2008" or "Things to watch for Next Year." Friend Cassandra had a go yesterday, wherein one of next year's expected trends was forecast as "disinvestment beats investment by a wide margin." In other words, everything's overvalued and cash is king.

While such a theory is easy for many observers to embrace in equities and credit, to date it has had little resonance for government bonds. Sure, govvys have backed up over the last few weeks, but in many developed countries yields are still perched very much to the low end of the 2007 range. The rationale, of course, is a "safe haven" bid amongst the sturm und drang of money market ruptures and equity volatility.

Yesterday Macro Man proffered the suggestion that nominal duration is an unttractive proposition; today he offers a bit of viiual evidence. The chart below shows real 10 year yields in the G4, deflated by the broadest measure of consumer prices, lagged one month. This means that that Gilts, for example, are deflated by the Retail Price Index (current reading 4.3% y/y) rather than the harmonized CPI (current reading 2.1%.) While the latter is the target used by the Bank of England, the former more accurately reflects the cost of living. Try, for example, to avoid paying your mortgage or your council tax bill by claiming that you "only consume the harmonized CPI basket", and see how far that gets you!
In any event, the chart shows that real yields have fallen across the G4, particularly in the US and Europe. Part of that is down to a base effect quirk in the CPI data, of course, but a good deal is down to the decline in nominal yields as well. In any event, real 10 year yields in the US are now not only negative, but at their lowest levels of the millennium. And all markets look much more expensively priced than at the beginning of the last global downturn. Judged from a real yield perspective, therefore, the investment prosepct for nominal government bonds bites.

What's particularly interesting is that the real yield gap between the US and Europe is now 1.37% in favour of the latter. This is the highest reading in favour of the EUR since March of 2001, a month in which EUR/USD traded through 0.90. While The Economist cover curse has done its work, generating a sharp correction lower in EUR/USD, Macro Man gets the sense that the market wants to make something more of the dollar. Could it be that people are actually turning....bullish?

On one measure, at least, so it would seem. The DB/Rusell Mellon position survey suggests a sharp increase in dollar longs at both the median and extreme levels. Such a move is difficult to reconcile with the trend in place until a couple of weeks ago, but might suggest that some deep-pocketed currency punters are beginning to use valuation as a more prominent investment metric.
Of course, that begs of the question of relative asset market valuation...which brings us back to the real yield measure noted above. Of the G4 currencies, US 10 year valuation is pretty clearly the least attractive. And since the US is still a "fixed income" currency, that should bode ill for the dollar in the medium term, barring either a bum-clenching recession (which shakes out long-term dollar shorts) or a sharp acceleration in growth (leading to higher nominal yields.) While each of these is certainly possible in 2008, neither is likely to emerge with certainty in the first quarter. Strangely enough, therefore, a bullish EUR/USD strategy into new year may have the twin attraction of being both fundamentally justified and contrarian!

One possible roadblock would be an unorthodox policy choice from the Middle East. What if, instead of allowing their currencies to appreciate against the dollar, they instead force the dollar to appreciate against other currencies? Forget the arguments on sovereignty and the appropriateness of foreign agents setting monetary conditions in the US; PBOC has been doing so for four years already.

But from the GCC's perspective, what would be easier? Breaking long-held pegs and switiching to a "complicated" basket....or just saying "200 billion EUR/USD yours"? It's not like they don't have the money to do so! And it would ease at least part of the problem, namely the decline in the trade-weighted foreign exchange value of the riyal, dirham, et al. It wouldn't of course, solve the issue of farcically low interest rates and the money creation resulting from partially sterlilized intervention...but then again these guys don't seem to care about that stuff anyway. Hell, they don't even seem to really care about the foreign exchange value of their currencies....why else would USD/SAR be at a multi-year in a week when millions of Muslims are convening on Mecca?

Finally, a bit of portfolio clean up. Macro Man will roll his short ESZ7 exposure to March this afternoon and will close the SGD swap overnight. Over the next week or so he'll look to roll out his put exposure on the Hang Seng as well. If Cassie and others are right and valuation is starting to matter, this is no time to be jettisoning short exposure to quasi-Chinese equities!


Charles Butler said...


What's curious, aside from the date, about the current continued volatility is that it seems also to be accompanied by a certain complacency. Some of the things that I follow which I would expect to be jittery under the circumstances are, in fact, not particularly.

The very real inflation story, though, is a case of bad news arriving when least able to be dealt with. One wonders if it foretells an episode of coalescence of sentiment in a US election year. (Tr., populist solutions to the fore.)

CDN Trader said...

We seem to be in the age of bubbles. There was equities, then housing....bonds seem to be a possible next candidate. If that happens, valuations do not really matter.

Brian said...

Hi Macro Man,

Many thanks for your previous link to alpha/beta theory. I realise your blog is aimed at already knowledgeable financial professionals but is there anywhere on you blog (or elsewhere) that might give a 'beginners introduction' to you your a/c jpeg e.g the open/closed trades etc.


Tradingmacroman said...

Macroman, nice blog, it's exactly what I'd like to do, I'm yet registered with a name very similar to yours... and we have the same job.
However sorry but after your reasoning on inflation and real yield (I agree with them..) I can't understand your position in TIPS, with a long 10yr TIPS you're locking a low real yield with an hedging on inflation, is it correct so that you're betting again on more FED easing??

Macro Man said...

Who knew my name would be so fashionable? Fabio, the TIPS bet is a bet on real inflation, i.e the compensation portion of the bond. Eventually, if/when things normalize, the nominal risk can be hedged away with short Treasuries.

CB, some element of protectionist ousturing seems unavoidable next year. If I were Hu Jintao, I'd buy myself a set of earplugs.

CDN, I think we have BEEN in a bond bubble, courtesy of Voldemort and other FX reserve managers. Now that even they have lost the appetite for mindless buying of Treasuries. If/when credit becomes slightly less sh1te, one would have to posit that bonds could/should be vulnerable, particularly at the front end...unless one buys into the "Fed cuts to 3%" theory, which I don't.

Brian, there's no real explanation that I've written. basically, the portfolio is divided between the simple quant beta strategies, and every thing else. Every month, the P/L is marked at the relevant fixing/closing price, which then becomes the opening price for the new month. Trades which are closed during the month get moved down to the 'closed trades' section, then removed once the new month starts.

Tradingmacroman said...

Thanks for your reply, what do you mean for real inflation?? Why not to trade directly inflation swaps??
Or do you play inflation as a good carry?

Mencius Moldbug said...


It is by no means a comment on your picks, which I'm sure are excellent. But I do wish people would stop using this word "real" as though it actually meant something.

When you bet on TIPS, you are betting on a bond whose yield is adjusted by a price index. This index could be the Dow, it could be the CRB, it could be the Baltic Dry. Instead it is the CPI as computed by some dude in an office in Rosslyn. Whatever.

The difference between the price of normal Treasuries and the price of TIPS reflects the market's current prediction of the future path of this index number. If the curve you expect to see this number follow differs from the market's current projection, and you are right, you can make money.

But this is no different from betting on Baltic Dry futures. There is really nothing special about the CPI. Which is why I find it confusing that this entirely different terminology, these words "real" and "inflation," are so often applied.

Again, though, I can't stress enough that this is not a substantive point. It is just a matter of economics and the English language - to paraphrase Orwell.

Macro Man said...

In fairness, the CPI is more comprehensive than something calculated by "some dude." Is it flawed? Absolutely. Is there a better measure of the cost of living? The Fed thinks the PCE, but that is possibly because it is marginally less volatile than the CPI.

The BDIY, among myriad other things, has a much more tenuous link with the cost of living, both in terms of scope and construction.

But there's no point decrying the word "real", since it conveys an idea that is understood by most financial market particpants and observers; to wit, a nominal (return, interest rate, whatever) deflated by SOME proxy for changes in the cost of living.

And yes, the return on TIPS is dicated by the evolution of a given proxy for changes of the cost of living- non-seasonally adjusted CPI. But insofar as language is meant to convey an idea, and most people understand what is meant to be conveyed by talking about "real" interest rates, I frankly don't see what the problem is.

Mencius Moldbug said...


The problem with the "real" terminology is that it implies the fallacy of objective value. While one can certainly use these locutions while constantly reminding oneself that there is no such thing as "value," only price, it creates extra mental work.

For example, it is commonly said that lenders demand higher interest rates to "compensate" for CPI inflation. This is simply nonsense. Lenders demand the highest interest rate they can get, and borrowers the lowest. There is no obvious, trivial linkage between the CPI and either the supply or the demand for future money.

Any more than there is an obvious, trivial linkage between the BDIY and interest rates. So, when we subtract the CPI from an interest rate and call that "real," we could do just the same thing with the BDIY. The truth is that we are combining two unrelated numbers into a single one, which is just bad math.

Worse, the second number is a pure fudge factor. There is no one "cost" of living. Different people buy different things all the time. The goods and services in the market are constantly changing. Even the indexes themselves change.

If you go over to John Williams' site, Shadow Government Statistics, he uses pre-Boskin-era techniques (or claims to) to calculate the CPI at 7.5% "inflation," not 4%.

If Williams is "right," "real" interest rates are negative. Very negative. But is Williams "right"? Or is the BLS "right"? The question has no answer. It is not an objective question. Whereas the current price of 10-year T-bills is very much an objective question.

So what you are doing when you talk about actual quantities of money as "nominal," and quantities adjusted by this index as "real," is deprecating hard numbers in favor of hard numbers plus a fudge factor. The number of ways in which this can mislead you is almost limitless.

The CPI, either Williams' version or the BLS version (they do tend to track each other) is not a bad way of tracking the prices of consumer goods. Perhaps it is a good predictor for, say, labor costs. But why should we care more about labor costs than about, say, shipping costs? Or raw materials costs? Etc.

Furthermore, the above does imply a specific problem with TIPS: as a TIPS investor, you are putting a lot of faith in Uncle Sam. As the results of the Boskin Commission show, Uncle has the motive, the opportunity, and the propensity to low-ball this number. Thus pulling money out of your pocket. The US is not Argentina, at least not yet, but...