First, an observation: yesterday's article, "Incredible Inflation Bond Bargain," received more hits than any other article I have written in recent memory. Apparently, people still are looking for bargains, and still looking for bond bargains as well. This is heartwarming to a bond guy, and of course even more to an inflation guy. But then, true bargains are rare, and true bargains offered by the government are even more rare. A belated hat tip to "Gratian", who asked me what I thought about I-bonds and provoked that article. Thanks for the suggestion!
There was some mild good news yesterday. Consumer Confidence rose more than expected, to 70.3 and only a couple of points below the post-Lehman highs set in early 2010 and 2011. Yes, 70.3 is still very low (the series is set so that confidence in 1985 equals 100, and in the recessions of the early 1980s and the early 1990s it was generally in the 55-80 range), but the longest journey begins with a single step. On the bright side, there's lots of room for improvement (see chart, source Bloomberg).
The internals of the Confidence number are not as good. Both "current conditions" and the 6-month ahead outlook improved, especially the outlook (when 'my guy,' whoever your guy happens to be, will be in the White House six months from now, surely things will be better), but the "Jobs Hard to Get" subindex, which is highly correlated with the level of the Unemployment Rate, barely nudged lower. Still, as depressing as it sounds, consumer confidence is a relative bright spot among recent data.
Home prices, as we have documented several times, are rising and the S&P/Case Shiller Home Price Index confirmed that by reaching the highest level it has seen since 2010. The 20-city composite is now rising at 1.2% year/year, which doesn't sound much but is the highest rate of change since the dead-cat bounce of 2010. Keep in mind that the index methodology involves a fair amount of smoothing, so it lags the actual improvement in the market. By comparison, the RadarLogic 28-day composite index as of the end of July recorded the highest year-on-year change since 2006 (see chart, source Bloomberg).
Also relatively good news was the Richmond Fed Manufacturing Index, which rose to +4 - not as good as it was earlier this year, but 23 points above its July low. The Richmond Fed district includes the "toss-up" battleground states of North Carolina and Virginia and the "leans Romney" state of South Carolina. It is encouraging that manufacturing in this region (with its 28 toss-up electoral votes) is outperforming activity in the Dallas Fed district (Texas, northern Louisiana, and southern New Mexico, none of which are considered toss-ups), the Chicago Fed District (which includes Michigan, most of Illinois and Wisconsin, and 6-electoral-vote-toss-up Iowa) and the Philly Fed district (which is Pennsylvania, NJ, and Delaware, and no toss-ups). This is merely an observation, and even if there were clear indications that the Administration was directing money towards projects in battleground states I wouldn't object to it - that's one of the prerogatives of incumbency. If you want that prerogative, work hard so that you can get to be the incumbent.
While the data points yesterday were good, stocks gave up the ghost and managed to lose most of the post-FOMC rally. That doesn't really shock me so much. Commodities, which should be more sensitive to inflationary monetary policy, are down outright since the Fed declared an unbounded easing policy, and both markets have rallied since June on the growing expectation of QE3. The fact that QE3 was larger than many observers expected caused some short-covering on the news, but I suspect most investors who thought QE3 was coming were already long their preferred assets. The actual open-ended Fed buying will definitely buoy commodities (which remain undervalued relative to past QEs) and might lift equities (which, however, offer fairly weak prospective real returns given the current market valuations), but we had already priced in some expectations.
And in the meantime, while yesterday's numbers were not bad, the overall picture remains pretty weak. I think the threat of sequestration at the end of the year will start to affect growth more seriously in October, because the end of the fiscal year for government expenditures is September 30th. Businesses that have the government as a significant client recognize that they may well be in Limbo on October 1st. This is what happens when government spending is 40% of GDP! The sequestration doesn't happen until January, so spending from October until December in theory will be unaffected. But, in practice, the government enters into contracts (for equipment and construction, for example) that cover many months, and it isn't entirely clear whether for example the Defense Department can enter into a one-year contract if it isn't known that the money will be there. I know several people in businesses that are directly affected by this issue, and they're concerned about it now, not just in January.
I saw an interesting study by State Street Global Advisors mentioned in a Pensions & Investing Online article. According to the study, about ¾ of institutional investor executives consider a 'tail-risk' event in the next twelve months to be likely. But here is the interesting paragraph in the P&I article:
"Survey respondents - money managers, family offices, consultants and private banks - expect the five most likely causes of a tail-risk event in the next year would be a global economic recession (36%); a recession in Europe (35%); the breakup of the eurozone (33%); Greece dropping the euro (29%); and a recession in the U.S. (21%). (Percentages total more than 100% because respondents could select multiple causes.)"
Apparently, 'inflation' isn't even on the radar as a tail risk. Of course, as an investor, what is more important than the tail risks you can estimate the probabilities of are the tail risks you aren't even thinking about or can't estimate the probabilities of. Incredibly, not only has the myth that recessions cause disinflation and deflation failed to weaken during the last few years, when weak growth has been accompanied by accelerating core inflation, it seems to have strengthened! While investors, as evidenced by the performance of inflation-linked bonds and of breakevens (and inflation swaps) and commodities, believe that inflation might well be a risk, it doesn't seem that many investors are focusing on it as a tail event. That is, they expect that a "bad" inflation outcome might be 2.5% or 3.0% core inflation. An outlier event to them may be 3.5% or 4.0%.
But what we know about inflationary outcomes is that if anything, they have tails that are quite long. And there's plausible reasoning which can produce very high numbers for that tail; see for example my article from late last month - before QE3 - called "What Keeps Me Awake At Night." I always take care to say that these concerns aren't predictions, but they are plausible possibilities, and the bottom line is that we don't really know how these relationships work at this scale. No central bank has ever dealt with numbers like this. It is a known unknown, and thus a source of a tail risk of indeterminate length.
In my opinion, when it's cheap to insure against such risks then it ought to be done. Presently, you can (as an institutional investor) protect against the risk that inflation will compound at greater than 4% for the next ten years for roughly 2.2% of the notional amount, or 22bps per annum. There are multiple ways to do this, some of which may be cheaper and all of which are beyond the scope of this article - but the point is that we have investors enumerating downward "tail risks" on growth while equity margins and valuations are high, and largely ignoring "tail risks" on inflation that could damage a number of different asset classes. I see lots of potentially dangerous scenarios for equities in October, several (but not all) of which are also dangerous for bonds.