More TIPS In Every Pot
Another slow day, with the equity bears/fixed-income bulls winning this time. The bond market ended up at 3.74% on 10y yields with TYM0 up 13.5/32nds. Stocks were roughly flat.
Inflation swaps rose slightly (1-2bps), despite falling yields. Forward inflation swaps continue to hover at levels around 3.0-3.2% for periods 5 years and out. It is surprising to me that inflation-linked bonds and inflation swaps are doing so well despite impending supply - tomorrow, the Treasury department will announce the size of Monday's 5y TIPS auction (more on that below), but there is more. The Wall Street Journal reported in this morning's edition (story for subscribers is here) that the Treasury is expected to announce in early May that it will increase issuance of TIPS in the second half of the year, potentially auctioning 10y TIPS every other month on a January-July cycle with two re-openings of each. This is welcome, and good strategy (as I argued in a column last December, if Treasury really wants to try and persuade buyers of nominal bonds that they are not planning to inflate away the debt, increasing the size of the TIPS issuance, rather than de-emphasizing it as they had been doing, has a great value in 'advertising' terms). And at these real yields, why shouldn't the Treasury issue as much as they possibly can, especially if they're not planning to monetize the debt?
Because I am often talking to people about inflation, I run into a lot of fans of gold who ask my opinion. My opinion is generally simple: gold does not pay a dividend, and its real return is approximately zero over time (you start with 10 ounces of gold, you end with 10 ounces of gold: real return=0%). While commodity indices offer interesting returns and an interesting pattern of returns, these come less from the movement of spot commodities and more from the other components of return: collateral return, risk premium, rebalancing return, convenience yield, and expectational variance.
Usually, people who like gold and own gold ETFs don't like to hear that it's a zero-real-return game (as an investor; of course, there are always profits to be had for successful speculators, balanced against the losses to be had for the unsuccessful), and sometimes they get sort of mad. So I am really happy to be able to present this link, from the Fama/French Forum. These two luminaries (Krugman has a Nobel, and Fama/French do not?!) answer the question of whether gold belongs in a portfolio. Their summary:
To summarize, gold makes sense as a portfolio asset only for investors who also get the consumption dividend from gold, since this "dividend" lower gold's expected capital gain. Thus, for investors who do not value the consumption dividend, the expected return on gold does not cover its risk as a portfolio asset, (which takes account of its value for portfolio diversification).
I had never thought about the convenience-yield angle. I am a little skeptical of the pure theoretical markets arguments that Fama/French crowd puts out sometimes, because the market isn't that efficient, but it's a point worth considering.
On Thursday, we will finally get some economic data. PPI (Consensus: +0.5%, +0.1% ex-food-and-energy) is not a market-mover since we have already had CPI last week, but Initial Claims could be interesting. After large surprises higher in the last two weeks, the consensus estimate is for a return to 450k (from 484k last week). But that's very timid, if in fact 450k or below is the true run-rate (two weeks ago, the forecasts were for 435k; last week they were for 440k), since the explanation for the prior misses - seasonal issues related to Easter week - should be smoothed out if we average over four or five weeks. If they are not, then that excuse doesn't really cut the mustard. With the last two weeks a cumulative 44,000 over the 450,000 that economists are now forecasting, we should be in the low-400s anyway. And if the true underlying pace of Claims is what they were forecasting before, 435-440k, then we could see a sub-400k print.
Personally, since confidence indicators have been weak lately I suspect the jobs picture is not improving as dramatically as economists have forecast, but either way this week's data point is important. If we get another 460k or 470k, it starts to look implausible that Easter could cause a three-week rise and we need to start wondering if perhaps the true pace of Claims isn't still around 460-470k, where it has been all year. It would not be the first time that economists had prematurely trumpeted a robust expansion.
Finally, at 10:00 Existing Home Sales (Consensus: 5.28mm) will be released. Recall that EHS had spiked on tax incentives to purchase homes; the thought seems to be that we will get another minor taste of that this month as those incentives are expiring. The recent volatility in this data series makes it almost possible to have a number that meaningfully changes our perceptions, unless it is very weak.
As I said earlier, the Treasury is also due to announce the size of the 5y TIPS auction, which will occur on Monday. The 5y auction is always subject to being a dog, but the level of real yields on TIPS suggests that the market could use a few more at the short end. I suspect the reason that the real curve is so steep, with real yields negative out for 3 years, is less because the market perceives expected growth as being negligible than because investors prefer to reduce duration risk until inflation is about to pick up. That is, with 20-year TIPS yields below 2%, a buyer of long TIPS will likely have a mark-to-market loss when inflation first picks up and all yields rise. Better, perhaps, to take a very low expected return (and negative in real space) with low duration risk, and then move out the curve once yields rise. It is basically the same dynamic that is causing such steepness in the nominal yield curve, although in the TIPS case the Fed doesn't have direct control because they set nominal, rather than real, interest rates. In any event, I think the unusual demand for short-duration TIPS right now will help the Treasury sell a lousy tenor at a lousy yield (roughly 0.55% right now in the 5y area).