Analyzing The Economic Conditions Needed For Persistently "Exceptionally" Low Rates
One of the key departures by the FOMC yesterday in comparison to prior rate statements, was providing a glimpse into what the specific economic conditions are that warrant continued "exceptionally" low rates. Among these the Fed outlined "low rates of resource utilization, subdued inflation trends, and stable inflation expectations"- so long as there is no perceived change to any of these, expect rates to persist in the 0-25 bps zone. Goldman provides some valuable insight on how to interpret these three key considerations by the Fed.
it is useful to consider what specific indicators FOMC members may have in mind as they consider whether “low rates of resource utilization, subdued inflation trends, and stable inflation expectations” remain in force. Current values of these indicators are shown in the table below, with the most important indicators (in our view) in bold text.
1. Low rates of resource utilization. The broadest measure of resource utilization is the output gap – the difference between actual and potential GDP. The Congressional Budget Office, the definitive source for estimates of potential GDP, currently estimates the gap at 6.3% as of the end of the third quarter, the largest since the early 1980s. (The gap shrank marginally in Q3, since the economy grew above trend last quarter.)
However, the output gap is a nebulous, difficult-to-measure concept and many analysts view it with skepticism. An alternative is to review more “‘micro” measures of resource utilization, a few of which are included in the table below. (A more extensive table is provided as Exhibit 5 of “Deflating Inflation Fears,” GS Global Economics Paper No. 190, September 29.) Of these, we view the unemployment rate as most important. It is a widely accepted surrogate for the GDP gap, it carries significant socioeconomic implications, and it is one of four indicators for which Fed officials provide public forecasts on a regular basis (the others are real GDP, core PCE inflation, and headline PCE inflation). At 9.8% as of September, the unemployment rate is nearly three standard deviations above its average level of the past half-century.
2. Subdued inflation trends. The use of the word “trends” in the statement hints at core inflation as a more prominent metric than headline, though a few FOMC members might take issue with that interpretation. Historically, current core inflation has tended to be a better predictor than current headline inflation of future headline inflation. In the short term, this is a distinction without a difference, as headline metrics show deflation over the past year and core inflation is on the soft side of desired levels. However, the distinction is likely to matter much more in 2010: headline inflation will rise well into positive territory in coming months as year-over-year comparisons become “easier”, while (after perhaps a slight increase in the remainder of 2009) we expect core inflation to decline further in 2010. The Fed forecasts price indexes for personal consumption expenditures (PCE) rather than the consumer price index, so we see the core PCE inflation metric as the most important.
3. Stable inflation expectations. A significant increase in inflation expectations could push the Fed to tighten even if actual growth or inflation remained on the weak side. In fact, a big move in inflation expectations in either direction would probably trump other indicators in terms of the influence on Fed policy. A case in point is the sharp decline in market-based inflation expectations in late 2008, which coincided with much more aggressive Fed rhetoric (including the statement that the FOMC would use “all available tools” to restore growth) and actions (the initiation of asset purchases in late November and expansion in March).
In our view, the most important indicators of inflation expectations are longer-term measures of breakeven inflation (the difference between the yields on nominal and inflation-protected Treasury securities; inflation swaps provide a slightly ‘purer’ alternative metric with fewer liquidity issues) and household expectations of inflation. In the case of breakeven inflation, the five-year, five-year forward measure is commonly used. In the case of household expectations, the longstanding University of Michigan measure of median 5-10 year inflation expectations is probably foremost.
After these two measures, the next is probably the long-term inflation expectations captured in the Philadephia Fed’s Survey of Professional Forecasters. We view forecasters’ expectations as slightly less important, as they generally lag a bit and the channels through which they affect actual inflation are less direct. In any case, longer-term market-based expectations are slightly above their average of the past few years while survey-based inflation expectations are slightly below. In short, inflation expectations are low and fairly stable at the moment.
Some Fed officials may also view the value of the US dollar or commodity prices (particularly gold) as containing information about inflation expectations, or about future inflation. It is more difficult to calibrate exactly how “out of whack” these variables are, and we very much doubt that they alone would prompt a change in Fed policy. However, large moves might be a contributing factor to a change in rhetoric if they accompanied modest changes in market- or survey-based measures of inflation expectations.
As for which of these three will show weakness first and provide at least some ammo for the Fed Hawks, here is how Goldman sees the sequence of priorities:
We see utilization as least likely to be an issue in 2010 – spare capacity is so large that it will take a long time to whittle away even with strong growth. In fact, where the unemployment rate is concerned, a simple rule of thumb known as Okun’s Law suggests that it could take many years to return to “normal” unemployment rates. (The rule is that two percentage points of above-trend growth, i.e. growth in the high 4% range, would reduce the unemployment rate by one percentage point per year.)
Core inflation is also highly persistent and we view the predominant forces as downward. Headline inflation, on the other hand, should perk up somewhat given commodity prices and easing year-ago comparisons. It is the most likely (although not very likely) to move above comfortable ranges in the next year, although as noted above, it is probably viewed by most FOMC members as secondary to core inflation.
Inflation expectations are the hardest variable to predict, and the process by which long-term expectations are formed is not especially well understood. Therefore, they have to be seen as a risk factor. However, insofar as some portion of expectations is “adaptive” (reacting to past inflation rather than predicting future inflation), lower inflation now would seem to point towards downside, rather than upside, risk to inflation expectations.
One thing is certain: the Fed will dither for years without doing much, even as another full blown liquidity bubble grows ever larger in its face. And by the time it wakes up and begins tightening, the market will likely again be in the parabolic phase last witnessed during the silly tech days of early 2000.