E-piphany

It was in general a “reboot” kind of day, two days before the Employment report. Overnight, equity futures launched higher; according to Bloomberg this was because the Chinese economy had some marginally stronger numbers. That seems to me unlikely in the extreme to have caused such a violent reaction. More plausible, but still of the wrong scale, was the report that Burger King is shopping itself (are they telling prospective partners that they can “have it your way”?) – recently, M&A activity has been more active than in some time, lending some sense of positive “creative destruction” going on.

But neither of those makes much sense in isolation as the reason for the strong overnight tone, and the concomitantly weak tone to fixed income.

The early data was slightly weak, with the ADP employment report showing -10k against expectations of +15k, but when the ISM Manufacturing report actually rose to 56.3 from 55.5, rather than falling as expected to 52.8, the play was writ. Bonds dropped off the shelf, and it has recently been a pretty high one, while stocks rocketed higher.

But again, the reaction to ISM was out of proportion to the usual reaction to ISM, especially when the surprising increase is along the lines of a wiggle on the chart rather than the sharp reversal that bond sellers seemed to have perceived (see Chart).

There was negative news on the economy, of course. What day would be complete without it? Car sales were again weak; the domestic sales at an 8.66mm unit pace represented the slowest August sales in a quarter century (although that statistic is aided by the fact that in August 2008 the crisis was only just becoming acute while in August 2009 the cash-for-clunkers program was peaking). ISM might be wiggling upward this month, but to steal Gertrude Stein’s line about Oakland, there’s no ‘there’ there.

None of this is cause, two days before Employment, for 10y yields to back up to 2.58%, for the stock market to catapult 3% higher and for the S&P to bust out of its “indecision zone” with feeling. Interestingly, despite the size of the price move market volumes were lower than on Tuesday’s relatively boring session.

The markets basically showed us that investors were over-short equities and over-long bonds, relative to their comfort zone, and it took very little to shake these weak positions out. This is good and bad news, for on the one hand the markets are now less proximate to breakout/breakdown points but on the other hand investors are also closer to neutrally positioned and hence a surprise on the Employment figure may have a larger effect than it otherwise would. But that is Friday’s news.

.

In yesterday’s column, I opined that the next truly big trade to come, in my opinion, is when both stocks and bonds decline as inflation expectations rise. I promised to talk about how I am positioned for such a possibility.

When I talk about that possibility, I am not referring to a one- or two-month phenomenon. I am talking about the reversal of the secular disinflation trend that we have enjoyed, more or less, for thirty years and which has been the prime driver in the stock market’s rise to exceptional multiples. Even now, after a decade of essentially sideways trade, the multiple on reported earnings for the S&P is still at 16.1.

Whether the 1970s inflation ends at the current level of rough price-stability or collapses further into deflation (I believe the former, of course, but if the latter then equity multiples also should suffer but high-quality bonds of course will do fine), the trend cannot continue forever. We have come very close to the limits of that trend. Sometime, and I think the process is already underway, we will enter a secular inflation trend.

I seriously doubt that we will have an inflationary trend that is long in duration; that is only feasible if the US currency itself enters into a lasting downward spiral. While not impossible, this is not my expectation. We can, however, have a general uptick in prices and a fairly brisk move to higher levels of inflation (even with a spike to, say, 8% before it settles back down some) that is global in nature and results from concerted cheapening of paper money relative to real assets as a way of effectively reducing global sovereign debt burdens. I think this is not only possible, it is likely and the main uncertainty is the timing.

In such an event, both bonds and stocks will suffer. Holding non-US assets is no panacea, because in a global inflation there is no reason the dollar would necessarily weaken – it might even strengthen.

Real assets will do well, but as we know the real yield on most physical commodities is zero (negative when you include insurance and warehousing) so if this isn’t a 2011 event but a 2012 or 2013 event, there will be a long period of weak performance.

If TIPS yields were in the 3-4% range they enjoyed in the depths of the crisis, the investment would be easy. If nominal yields were high, then coupons would give some protection against the capital loss caused by higher yields. If dividend yields were high, then there would be some margin of safety protecting equity claims against a cheapening of multiples. Unfortunately, none of these things is true.

In my view, investing for the big turn is about treading water, taking only such small risks as the market pays us to take – although almost no risk today carries much compensation – and for which the likelihood of serious debacle is low.

When I look at the fixed-income part of the portfolio, I want to stay short duration of course because I think yields will be rising at some point in the not-too-distant future. The fact that other people believe this also is one reason the yield curve is quite steep and short yields are very low, because many investors are living in the very front end of the curve. Tbills in normal times are a good safe investment that even tends to keep up with inflation – but these are not normal times; central banks are keeping short rates ultra-low and if I roll TBills at 0.15% (and they could go negative again if the FOMC institutes a penalty IOER as I suggested yesterday), I will get nothing, or less than nothing after inflation, for perhaps several years – even if inflation rises sooner than that. I need to go out the curve a little bit, but how far?

One way to look at this problem is to ask “how much am I being compensated for each additional year of maturity?”[1] The two charts below (Source: Enduring Investments) show the current spot Treasury curves for nominal interest rates (first chart) and real rates (second chart) in red, along with the curve of one-year forward rates in blue. That is, each point on the blue curves answers the question “how much am I being compensated for a one year extension starting from year x to year x+1?”

These charts tell me that I might want to go out as far as 5 or 6 years on the curve, but no further as the years after that point keep adding risk but don’t add much return. Moreover, I can compare the real and nominal forward curves to see how much fixed yield I am giving up for the inflation-linked part of the TIPS return. The chart below (Source: Enduring Investments) shows the difference between the blue lines on the prior two charts.

I am willing to give up ½% or so over the next year in order to get actual inflation. I am willing to give up about 1% over the following year (the first point on the chart above is the 1y, 1y forward) to get inflation in that second year. And in general I think that trade of fixed yield for inflation possibilities is a good deal for the next couple of years after that.

The consequence of this analysis is that in fixed income, I wouldn’t hold nominal Treasuries at all. I would hold 4-6yr TIPS. I prefer the July-14, July-15, and July-16 issues. Holding a TIPS ETF or a mutual fund is not my druthers, because an ETF will hold all of the bonds in the index and both an ETF and a mutual fund will maintain their durations rather than “rolling down the curve.”

Although the real yield for these bonds is in the 0.11%-0.42% range, it’s better than nothing while I wait to reinvest at higher yields. That is, of course, the plan: when yields rise, it won’t hurt my portfolio very much because the maturities are short, and in a few years there will be ample opportunities to invest at higher yields. I’m treading water while I wait, and locking in a positive after-inflation return. (Similar analysis can be performed on non-US curves, for those who have the ability to invest in non-dollar bonds).

I want some shorter-duration stuff that is still protected against inflation, and with short real yields negative, there is room in my portfolio for those zero-real-yield commodities. However, I prefer commodity indices, which have other sources of return besides the return to the spot commodity. I own GSG, an ETF that tracks the S&P-GSCI index, but am considering changing the position to UCD, a leveraged ETF that tracks (2x) the DJ-AIG index. Of course, while this is “short-duration” in a fixed-income sense, commodity index investments are very volatile.

For diversification, I do hold some equity positions, but these I want to be as bondlike as possible. I seek high dividend-payers with low leverage. I still expect these to lose in the next leg of the bear market, but again my purpose here is to earn something in the “shoulder period” while the secular trends are reversing and to avoid being carried out when the first waves hit. I am fixed-income heavy. In addition to the TIPS mentioned above, I own OSM, an inflation-linked exchange-traded note issued by Sallie Mae that I wrote about in this space back in May (link to comment).

This is not a portfolio that will shoot the lights out, but then again it is also a portfolio that won’t get shot at very much, either. My philosophy is that I have worked hard for my assets, and I am primarily concerned with keeping them (and, of course, maintaining their real value)…especially in times like these, when the downside risks are legion and the upside opportunities rather scarce. It won’t always be like that, but for now boring is beautiful.

---

[1] Technical note: Fixed-income purists may object that we should more-precisely look at the added compensation for each additional unit of risk, a concept that explains some of the curvature of the spot curve. That is, the extension from a 10-year maturity to an 11-year maturity adds much less duration than the extension from a 1-year to a 2-year maturity. However, at interest rates this low the effect is actually fairly small over the interval of the curve that is of interest, and using maturities rather than durations doesn’t affect my answer.