Weekly Wrap-up: Employment and Inflation - A Correlation?
Monday, August 07, 2006
Last week in this space we examined the tenuous statistical footing on which stands the view that economic growth and inflation are positively correlated. They're not. (Please examine last week's article for the charts and discussion on this topic.)
This past week the stock market was atwitter in anticipation of the Nonfarm Payrolls data, under the earthly illusion that a weak payrolls number would guide the Fed in deciding whether or not to pause in its rate-hike trek. The thinking is, I suppose, that if jobs growth is weak, then inflationary pressures are less than if jobs growth is strong, and in light of weak Payrolls numbers the Fed would conclude that growth (and therefore inflation) would decelerate.
Once again, rather than accept the CONVENTIONAL WISDOM, we figured that we should have a look at the data.
This first chart plots the Y/Y change in the PCE Deflator against the Y/Y change in Nonfarm Payrolls. (We're using the Headline PCE Deflator rather than the Core PCE Deflator for the reasons we discussed last week, that with energy prices protractedly elevated rather than simply seasonally volatile, it makes less than no sense believe that the Core number is a more accurate predictor of future prices than are the Headline number.)
What struck me about this chart was the independence of the cyclicality of the Nonfarm Payrolls Y/Y Change (red line) from the Inflation (the blue line). Arguably there's a 4 or 8-year cycle on the red line that's quite pronounced and that makes little or no impression on the PCE Deflator. And indeed over this 46-year period there is an essentially insignificant -0.05 correlation between these two series.
There are brief periods of strong positive correlation, such as the 2000-2004 period highlighted above, but, for instance, over the past 15 months the relationship between the 2 series has reversed itself to a modestly inverse -0.22.
It appears clear that whatever correlations in the 2 series do show themselves for brief periods are at best extremely unreliable. And the M/M data is, if anything, noisier and less reliable.
Now, we did find that if we set the PCE Deflator Y/Y Change (blue line) back by 20 months we got a somewhat more satisfying +0.32 correlation in the 2 series.
With the blue line pushed back, 20 months on this chart, there is an increase in the periods of strong positive correlation (highlighted in yellow). But that said, the overall correlation is still just modest and no better. (And since the early '80s the correlation of the 2 series on this chart is also very close to zero.)
Now let's think about what this means. Slowing jobs growth does tend to suggest (even if only in a meek voice) that 20 months hence inflation is likely to decelerate.
Now, the question we have to ask ourselves is whether we really think that the Fed's concern at this point should be inflation 20 months hence, or if they shouldn't be, with oil trading over $76 and Gold near $650, more focused on what the numbers are going to look like BEFORE the spring of 2008.
Once again, the evidence for the Conventional Wisdom (in this case that slowing jobs growth will diminish inflationary pressures in the US Economy) is simply nonexistent in real time. And while there is some modest evidence that inflation may taper off in a year-and-a-half or more, the case this data makes is not particularly strong, nor is the prediction it makes close enough at hand that it should guide Fed policy. (How reassured would you be by the Fed saying, "Don't worry, inflation will calm down by the spring of '08?")
WILL the Fed pause on Tuesday? The futures markets are saying that they probably will. SHOULD they pause? Neither slowing economic growth nor slowing Payrolls growth suggest that inflation is about to decelerate.
As Robert DeNiro's character, Michael, so famously said about a bullet in The Deer Hunter, "This is this. This ain't something else. This is this." The same is true about inflation. It ain't something else (economic growth and/or Payrolls growth). Inflation is inflation. It's rising. And we haven't found anything but Conventional Wisdom to suggest that it's on the verge of doing anything but continuing to do so.
*** *** ***
The SPX has completed Trading Day #961 of its 4-Yr (1008-day) Cycle.
Since last December we have been expecting this 4-Yr Cycle to play out fairly benignly, roughly along the lines of the 1990-94 cycle (royal blue line). We have worried about the possibility of a more malign scenario, more like 1966 or 1990, but at this point, with the SPX tracking the 1978 (electric blue) performance, the case for a benign 4-Yr Cycle Low appears intact.
Aggregate Earnings Estimates continue to trend higher.
The quality of earnings remains solid as there is no sharp widening of the gap between Reported Earnings (pink line) and Operating Earnings (yellow line). Growth prospects continue to be robust as represented by the expected +10.3% growth in earnings over the coming 52 weeks (91.26 / 82.76- 1 = 10.3%).
The difficulty for the broad market continues to be the distribution of earnings, with the largest plurality of growth coming in the Energy patch.
Relative to July 15, 2005 the SPX shows Trailing 52-Week Operating EPS up +17.5%. However, the SPX Ex-Energy shows EPS up just +10.5%. And as you can see, the spread between these two figures has begun to widen again.
Relative to one year ago the SPX shows Trailing EPS up more than +15%, but the SPX Ex-Energy shows EPS up just +9% over the same period.
The SPX Ex-Energy's Y/Y growth has dropped out of double-digit territory while Energy stocks are showing >40%+ trailing EPS growth. The way growth is distributing itself continues to be inflationary, with Energy and Materials prices continuing to raise input prices. And we continue to doubt that the table is yet set for a new bull market.
The function of cyclical bear markets, among other things, is to flush excesses of speculation and liquidity. And those excesses have not yet been flushed.
*** *** ***
Much has been made of the 10-Yr Treasury's drop from 5.2% back down under 5% since the beginning of the summer. Part of what's happened as a result of that decline is that the Yield Curve (defined here as the difference between the 10-Yr Treasury Yield and the Fed Funds Rate) has inverted even further.
The SPX Price/Earnings Ratio on Forward 52-Week EPS, now at 14, will continue to have trouble expanding (and may even fall farther) unless/until the Yield Curve begins to flatten out and move from "upside down" to "rightside up."
The 10-Yr Treasury now yields LESS than the 3-month annualized Headline PCE Deflator (+5%) and just 1.7% more than the Y/Y PCE Deflator. The bond market is effectively making a huge bet that Nominal GDP Growth (the sum of inflation and real growth, most recently +6.87%) will slow drastically, by about -2%.
If Nominal GDP growth slows to +4.9% (as one could argue that the bond market is suggesting) with, say, +3% inflation, that leaves just +1.9% in Real GDP growth. But in a +1.9% GDP-growth environment, it's going to be awfully tough for EPS to grow by +10%. If, on the other hand, the Inflation Doves are correct and the PCE Deflator drops down to, say +1.9% (inside the Fed's comfort zone), then near-trend +3% growth makes EPS growth a lot easier.
The question is, if the Fed pauses, do we really think that inflation is going to DROP? How about if Crude Oil breaks above $80 and Gold breaks $700 to the upside? The deeper question may be whether the Bernanke Fed is really rigorous in their commitment to controlling burgeoning inflation or if they're going to rely on Conventional Wisdom that has little or no basis in the data.
We'll know a lot more (well, maybe only a little more) after the Fed meeting on Tuesday.
Best regards and good trading!