Paperback Writer

Over the weekend, Barron's published an article singing the praises of adjustable-rate preferreds. As is typical, Barron's trucked out Bill Gross to lend legitimacy to its argument. The article quotes Mr. Gross's opinion that recent government rescue measures for banks makes both bonds and preferreds issued by these institutions safer. He may be right about financial sector debt, at least out to June of 2012 (the end of the TLGP FDIC program - I wouldn't trust any financial bonds other than those issued by the biggest deposit taking banks which did not need extra help from the government beyond that point), but I think preferreds, especially traditional non-cumulative preferreds are more at risk for dividend suspensions. Why? because the cash saved can help a bank's capital ratios without further government assistance. We have already seen the wiping out of GSE preferred dividends and the Fed has asked several regional banks to suspend dividend payments. Citing the fact that a certain large bank has essentially wiped out its common dividend, but has left preferred dividends untouched as a point of comfort is foolish. By doing so this bank has made the suspension of preferred dividends the next line of defense. The 200 to 300 basis point spread between this bank's non-cumulative preferreds and its cumulative trust preferreds says that the market is concerned. But dividend suspensions are not the crux of my critique of the Barron's article.

My criticism centers on the relative value of adjustable-rate preferreds. The article states that adjustable-rate preferred dividends can only go up from this point. This is generally true. However, this does not necessarily mean that they will experience price appreciation. Why? Preferreds are credit products which trade at spreads versus treasuries, but which treasuries?

Preferreds are long-term or, in the case of non-cumulative preferreds. perpetual securities. This means that they trade versus long-dated treasuries. However, their coupons are set versus short-term benchmarks (usually three-month LIBOR). This means that one doesn't necessarily need short-term yields to rise in absolute terms, but rather in relation to long-term yields. LIBOR-based preferreds perform best perform best when the yield curve is flat or inverted. We have evidence of this with MERprG. MERprG reached a premium of $26.55 on 2/2/07. The yield curve at that time was slightly inverted. I don't think we will see economic conditions necessary for a curve for a long time.

The Fed will do what it can to keep short-term rates low for the time being, but even when the Fed feels the need to raise short-term rates (during an eventual expansion) inflation caused by not only growth, but the massive amounts of money being printed (not to mention the huge quantities of debt being issued) promises to push long-term rates higher as well. This means that even if credit spreads between preferreds of any kind and treasuries tighten, prices may not rise much, if at all, because the coupon on the floaters will lag long-term rates, I.E. the yields at which new preferreds would be issued. This would keep the floaters priced at significant discounts.

Floaters may not trade any better than existing high-coupon fixed-rate preferreds. Most floating-rate preferreds spread their coupons anywhere from 35 basis points to 100 basis points over three-month LIBOR. Thus, even when considering a preferred which sets its dividend 100 basis points over three-month LIBOR, we would need to see three-month LIBOR at 7.00% or more for a 100 basis point floater to equal the coupon of existing high-coupon preferreds. The last time we had three-month even at 6.50% was after Alan Greenspan jacked the Fed Funds rate to over 6.00% quash the tech bubble. One also must consider that any price gains one experiences on floater may be similar to those on high-coupon preferreds and, even if they are greater, one has to consider the income deficiency of the floater versus the fixed-rate. Total return still probably favors the high-coupon fixed-rate preferred.

Lastly the article describes floating rate preferreds as a hedge versus inflation. This only exposes the author as being unknowledgeable. Long-term rates reflect inflation expectations, not short-term rates. If higher inflation is expected, these preferreds should trade in ways which reflect a steeper yield curve. TIPs five to ten years out are cheap. If one wants an inflation play, buy TIPs. Buying floating-rate preferreds to play inflation is like buying a minivan to go racing at Indy.

Have a Happy New Year!