All of the discussion about tapering has long since moved beyond the "if" stage and has been centered on the timing dimension, treating the tapering as a virtual certainty within a relatively narrowly constrained time horizon which is become the focus of the debate. Any remaining commentary to the effect that tapering may not occur in the foreseeable future has long since dried up and blown away. Reading the various speeches of Fed governors is generally supportive of the idea that this is a question that is solidly in the "when" camp and not in the "if" camp. Even St. Louis Fed President Bullard, who argues convincingly in favor of waiting to taper until economic data of sufficient strength to justify it becomes visible to the naked eye; acknowledges that he expects to see that level of strength in the data. The "No Taper" constituency, for the moment anyway, is a small and non-vocal one, if it exists at all. The October jobs report, at 204,000 dramatically better than the 120,000 expectation, should close out any holdouts on the inevitability of the phase out of quantitative easing.
With the expectation of tapering as a done deal, the topic of debate is all about when it will occur. The most obvious answer is the March meeting when Chairperson Yellen presides at her first meeting. The strength of the employment report for October, however, will amp up the volume of those who expect a tapering action to be announced in December, or January. There is no Fed meeting in February and any intermeeting announcement is very unlikely. There had been folks calling for the late taper, out in June or July of 2014. We could see those forecasts come in a few months now that employment is strangely back on track, averaging 200,000 per month over the last three months courtesy of the strength of the October report and revisions to previous months, after the melancholy over the last few months labor reports and the presumed weakening economy.
What is curious about this whole state of affairs is that the market is regarding a reduction in bond buying by the Federal Reserve as a virtually certainty, expecting that it will fall within a fairly tight range with respect to the amount of reduced purchasing, and that will occur against a time frame that is fairly tightly bounded. That being the case, this is a Finance 101 discounting problem waiting to be solved!
The odd logic of the Treasury market has the 10-year currently yielding 2.75% after having peaked at close to 3% in early September. The October employment report will fan the flames of those advocating an early commencement to the taper. Given that the taper is viewed as a lock anyway, and there is little time value to money given the extremely low levels of short-term interest rates, why doesn't the market just fully discount the tapering reality now and move yields on 10-year notes to somewhere north of 3.00%? Isn't this the way we are taught that markets work?
We are taught to take future cash flows, weight them by probability, and discount them at the risk-free rate to get a net present value. The math seems to suggest that if you take a 3.00% yield in March of 2014 on a Treasury 10-year, convert it into a dollar price, give it a 100% probability of occurring, and then discount it back to today using something very close to zero; what you end up is with a 10-year note at a dollar price that converts to yield very close to 3.00% (that's not a surprise). Yet, despite the significant sell-off after the October employment report, we are not there. Not yet anyway.
The rational investor would sell now if he knew with certainty that rates would be higher at some point in the next two to six months. Yet the sellers have not materialized sufficiently to drive yields back to the 3% level they reached in early September just before the taper-train ran off the tracks. If investors are rational, given today's yields on Treasury securities, then they must believe that yields in September over-discounted the impact on the market of the lost buying power of the Federal Reserve on the Treasury market. Or, alternatively, that the market was assuming too rapid a decline in the amount of the taper.
Another alternative explanation is that the market may in fact be suggesting that tapering is not a done deal after all. March is still a long time away and there are many things could happen to economic performance that might extend the taper date even further, perhaps forever, closing out the discussion entirely. Maybe the market simply over-reacted, producing the 2.98% peak in 10-year Treasury yields in early September, when in reality the taper should produce yields closer to where we are now. Perhaps now that we are going through this process once again, the market may be valuing the impact of the taper in a slightly more restrained fashion, or more "accurately". Or maybe the market really doesn't know what the impact of a taper is actually worth on bond yields after all?
My question is, if the market is struggling so much with anticipating the market impact of an event with probability of 1.0, a 2 to 6-month horizon, and with a discount rate of almost zero; how is it going to figure out where a 10-year should be when the Fed decides to withdraw from the bond market completely?