I had asked David over at The Aleph Blog, one of the many fine websites on my links page, about total return on a risk-adjusted basis for 10-Year Treasuries bought during the last “stagflation.” He had made a comment that prompted my question:

During the last Stagflation, bonds were called “certificates of confiscation” by many professionals in fixed income. It paid to have your fixed income assets as short as possible.

I can understand that perfectly from the standpoint of an institution that needs to avoid showing a paper loss on the holding value of the bond as yields rise and price falls. However, I don’t understand it from an individual’s standpoint, especially if that individual has a nodding acquaintance with the Austrian Business Cycle Theory (ABCT).

There approaches a point where even holding to maturity and shoving the proceeds from coupons under your mattress begins to make sense, and at that point, I think one holds one’s nose and buys the damn bonds. From Jan-76 to Dec-80, yields on the 10-Year Treasury ranged from 6.8% to 13.65%, with an average and median of 8.87% and 8.39%, respectively. About 1/4 of the time, yields were double-digit (291 of 1263 days, visit my links page and say “hi” to FRED II). A 10% coupon annually, even if the proceeds are shoved under your mattress, doubles your money in a decade, and the “rule of 72″ says that is 7.2% annualized, although Excel says that’s 7.1773% annualized. Any guesses on the long-term historic yield on dividend-reinvested stocks? Is 7.2% annualized not approaching a risk-adjusted “time to buy lots of bonds” situation? Reinvesting a 10% annual coupon from a 10-year Treasury into something like 1-month CDs at 5% (not unreasonable if 10YT is 10%) gives an annualized 10-year gain of 8.48%. With yield in the teens (at 13% even), the annualized gains are 8.7% (coupon payments under the mattress) and 10.2% (coupon payments to 5% CDs). Ahem!

Back to Austrian Business Cycle Theory (ABCT). Every fiat currency tightening cycle ends with a loosening cycle and another credit bubble in some asset or another, usually some financial asset because of the Cantillon effect.

What would the results have been, if one had, in the late 1970s, bought long- or intermediate-term bonds to hold to expiration, and bought another asset with the coupons? It stands to reason that real estate shouldn’t have been the chosen asset, as it was already in a bubble in the late 1970s, so one should have looked for an asset other than housing. Like maybe bonds or stocks? How would that have worked?

Actually, I covered that line of thinking in July. See the fourth chart down, at the point of double-digit yields, diminishing returns start to show up in subsequent stock gains, meaning that taking the safe money (10%+ coupons) and finding a home for the payments is the way to go.

I’m best served by not thinking like an institution. When, if ever, interest rates get to the point where non-reinvested coupon payments approach the historic long-term returns of stocks, and rates are still rising, I won’t be thinking of bonds as “certificates of confiscation” and I won’t be looking at the short end of the curve. Except, maybe, thinking of the short end as a place to park coupon payments from the long end, while I try to figure out what bubble will get inflated once the loosening starts. Because as long as there’s a Central Bank, it will start …