E-piphany

The world’s central bankers seem to be aware, suddenly, that the global economy isn’t doing great. Where they’ve been for the last year is hard to tell, but abruptly they seem to be getting religion on the topic of sluggish economic growth. It is ironic that this happens right after the NBER declared the end of the recession, but better late than never I suppose.

This comes as signs of an imminent second dip (if you want to call it that), that were epidemic early in August, have faded. If I were inclined to be generous, I might suppose that the recent acceleration in talk about further monetary stimulus is related to the fact that it has only recently become apparent that not only is no fiscal stimulus is riding to the rescue any time soon, there is also a chilling wind of fiscal restraint starting to blow.

The tenor of comments has recently begun to tilt decidedly in favor of further monetary action, to the point where it almost seems as if the intervening data over the next month or two won’t matter unless it is really, really strong. Last night, minutes after I posted my commentary, the Nikkei reported that the Bank of Japan is considering buying securities backed by small business lending, and increase purchases of government bonds. I have said before that Japan’s experience with persistent deflation is more to a lack of will than to a lack of a solution (see my comment here).  Maybe they’ve decided to step up the dosage.

At the same time, we are getting increasingly convergent comments from Fed speakers. Today Chicago Fed President Evans, in an interview with the Wall Street Journal (link) said that unemployment is not falling “as quickly as it should” and the Fed should deploy “much more” monetary accommodation. Evans’ argument, by now becoming familiar, is that the Fed needs to push inflation expectations higher so that real rates can be lower than zero while nominal rates are bounded by zero. (The Fisher equation says nominal rates ≈ real rates + expected inflation, so if nominal rates are at zero the only way for real rates to become more negative is to raise expected inflation.) Now, ideally the Fed could push inflation expectations higher without actually pushing inflation higher, but because the Fed is transparent and investors (and consumers) can read, they stand no chance of fooling the market by simply talking up expectations. They need to take concrete steps to force inflation higher; this will also raise inflation expectations.

Ordinarily, the Fed would have difficulty doing this because pushing inflation higher is normally against one of the Fed’s mandates-in-twain. When the Fed pushes, growth goes higher but inflation goes higher; when the Fed pulls, growth and inflation go lower. But in this case, with inflation lower than the level the Fed considers is consistent with “stable prices,” they can pursue an accommodative policy single-mindedly without worrying that one of their mandates is being sacrificed to the other. (To be sure, this is mostly semantics. The FOMC would love to overshoot the 1.5%-2.0% target right now, but if inflation gets away a little bit it must appear to be accidental).

Unfortunately, buying more securities in QE2 is unlikely to push prices much higher as the program is currently constructed, because the continuation of payment of interest on excess reserves (IOER) will cause the reserves to mostly stay on the balance sheets of banks and Fed speakers have barely mentioned IOER. At the margin, some money will flow into the economy, but this delta is indeterminate and surely a small fraction of the amount of the purchases.

Still, the inflation risks are clearly to the upside, especially since (aside from housing) prices are already rising at a 2-3% pace, and forewarned is forearmed. Inflation markets themselves do not yet fully believe it, but the trend is improving. The chart below (Source: Enduring Investments) shows the front of the inflation swaps curve, where we would expect QE to be most effective. The inflation swap curve has been steadily upward sloping, so over the last year the lowest projected inflation has been for the ensuing year, with slightly higher inflation in year 2 and inflation higher still (but still low!) in year 3).

Inflation expectations are rising, but not exactly spiking.

You can see that the recent talk about QE2 has lent a weak bid to the inflation market, but even 2-3 years out – the top line covers inflation from October 2012-October 2013 and is only a smidge above 1.5%. While longer-dated inflation has risen a bit (10y CPI swaps are 30bps above their lows), this is evidently not coming as much from the front end as I would expect, and that means it isn’t a pure response to the reasonably quick-hitting nature of QE2. Perhaps investors are seeking to add inflation duration while they still can do so at reasonable prices.

Now, if you’re a retail investor you can’t invest in inflation swaps and so add pure inflation duration. At this time, the universe of financial products – especially retail financial products – that can be plausibly claimed to be inflation-linked is scandalously small. After a number of TIPS mutual funds (around 40 by my count), most of the other products can best be described, and generously at that, as somewhat inflation-related. Included in this latter list are physical commodity ETFs, commodity indices, residential real estate, commercial real estate, REITs, Timber, and infrastructure, as well as some products in the mutual fund space that are generally mostly equities: HAP, GDX, MOO, and so on.

Here is why it matters. Let’s suppose that you want to invest in a floating-rate note tied to LIBOR, on the theory that short rates are highly correlated with inflation (as they are, in fact) and so should provide some inflation protection. The chart below shows the results of a little simulation I did to illustrate this point. I simulated a path for inflation and a path for LIBOR consistent with a correlation of 0.8 between LIBOR and CPI and several other assumptions that aren’t strictly important for understanding the result. For each of 250 simulations, I calculated the IRR of a 10-year floating-rate bond that pays L+100 and the IRR of a 10-year TIPS-style bond that has a 1.5% real coupon. The chart below shows the scatterplot of inflation-linked bond returns (ILB IRR) versus LIBOR-based bond returns (LIBOR Bond IRR), along with the R² of 0.6 which confirms that the resultant correlation is approximately 0.8 (the square root of 0.6 being about 0.77). This would seem to suggest that a LIBOR-based bond is a reasonable substitute for an inflation-linked bond if the correlation between short rates and inflation is reasonably high.

However, this would not be entirely correct! The nominal return is not, strictly speaking, what I care about as an investor. I want to maximize my real return, not my nominal return. And performing the same exercise with real IRRs gives a very different picture. In real space, the volatility of the real return of a held-to-maturity TIPS-style bond is approximately zero, as you earn the stated yield (in this case, the 1.5% coupon on the original par) regardless of the inflation outcome. But the ex-post real return of the LIBOR-based bond is still quite variable as long as the correlation between inflation and LIBOR is not 1.0.

The implication is that investments which merely proxy for inflation and do not explicitly return CPI like inflation-linked bonds and swaps do will provide significantly more real volatility than do inflation-linked bonds and, consequently, must offer significantly more prospective yield to be viable alternatives.

This doesn’t mean that TIPS at -0.5% real yield are a great investment, but it means we need to be cognizant that the extra risk we are adding when we go to an investment that isn’t explicitly indexed is a lot higher than we might think, if we just look at nominal correlations.

And this takes us to equities, commodities, and all of the other supposedly inflation-related markets that blasted off today (along with TIPS) given the appearance of central bank synchronicity around the issue of quantitative easing. Stocks ripped 2.1% higher, on improved volume for a change. Crude oil, corn, sugar, gold, silver, coffee, beans, and wheat were all up more than 1.5%. The dollar was weaker, even against the yen. This latter point makes little sense: inflation which is idiosyncratically ours, or profligate monetary policy executed by our central bank compared to other central banks, ought to weaken the dollar. But concerted profligacy or broad global inflation shouldn’t affect the unit, should it? The dollar, especially, looks like a knee-jerk response, and other markets are probably also overreacting.

.

In other news, following up on Friday’s comment: the Treasury today reported that while they made a cool $1bln on the Citi sale, they expect to lose money on the housing programs and auto rescues within TARP but expect net TARP losses to be about $50bln. The table below, from page two of the Treasury’s report gives the breakdown.

Now, the Treasury also goes on to say that they expect “substantial losses” on Fannie Mae and Freddie Mac, although those stem from decisions made before the conservatorship rather than after. The report attempts to offset this against the “substantial gains” made by the Fed from their purchase of $200bln mortgage securities; of course, those gains (how much is of course highly variable and dependent on continued low rates and a downside surprise in default experience) are only gains if the securities are marked at the current market price rather than their expected sale price! But the bottom, bottom line remains: the government will have bled away hundreds of billions of dollars to keep the unemployment rate down around 10%.

In economic news, the Non-Manufacturing ISM today came out roughly on expectations. The Prices subcomponent was roughly unchanged from the prior month, around 60, in contrast to the sharp rise in the ISM Prices subindex. So, the jump in the latter is either something particular to manufacturing, or it is largely spurious. I expect the latter.

And finally, lest you think that you have problems: Jerome Kerviel, the “trader” who circumvented internal controls at Soc Gen to run huge (and money-losing) positions, was ordered by a French court to repay the bank €4.9 billion. It will be hard for him to earn the money during the next three years, as he will be in prison. I’ll bet Kerviel is thankful for low rates, since even at a 1% interest rate he would be on the hook for €49mm in interest every year.

Tomorrow, the ADP Employment Index (Consensus: 20k vs -10k) is expected to show a tiny uptick in jobs. Because of the Census additions and now subtractions, this report has gotten less and less airtime but it is worth keeping an eye on – the market will react to outliers.