Five economic charts

One of the nice things about regular exercise is that it tends to free up the subconscious to wander where it will as one labors through strenuous physical exertion.  This is particularly true of Macro Man's sport of choice, road cycling, which is synonymous with suffering (OK, and performance-enhancing drugs as well, both those just mean that the guys who take them can suffer for longer.)

Of course, one of the downsides of allowing your subconscious to wander on its own is that there's no guarantee where it will end up.  Macro Man is conscious that he's spent a lot of time on the US recently, while there are other issues- Canadian elections, the forthcoming Chinese Plenum, this week's ECB meeting, to name a few- that merit comment in their own right.

Yet as he finished climbing a hill in the cold wind yesterday, Macro Man found his mind returning to the US and a few issues relating to some of last week's posts.  Probably the best way to express these thoughts is through a few charts:

*  There has been a lot of discussion about economic models and policy rules, unemployment, underemployment, labor slack, etc.  What's indisputable is that the Fed has consistently over-estimated the unemployment rate.   The chart below calculates a rolling 12-month forward unemployment rate forecast from the SEP projections and compares it with the actual unemployment rate 12 months after the forecast.  Readers will be unsurprised to see the consistent miss:

*  Of course, this misses the whole issue with the participation rate.   Macro Man's insight while riding yesterday is that we could turn these forecasts on their head, rendering them as employment rate forecasts.  This in turn allows us to plug in a number of participation rate assumptions to render a spectrum of forecasts to compare with the actual employment level out-turn.   The chart below shows the actual level of employment 12 months after the forecast date in dark blue, and three different forecasts based on different participation rate assumptions. 

The light blue line uses the actual participation rate at the time of the relevant employment release (i.e., 12 months after the forecast.)   Unsurprisingly, this is essentially the same relationship to the actual employment level as the unemployment rate forecasts in the chart above.   The red line assumes that the participation rate stays static for 12 months from the time of each forecast period (rather then declining, which has been the trend.)  As you can see, actual employment mostly undershot the implicit Fed forecast in 2011-12 before generally out-performing since early 2013.   Perhaps uncoincidentally, the Fed first mooted the taper soon after.  Finally, the purple line assumes that the participation rate will converge on 65% over a three year horizon; in other words, this forecast projects that the participation rate will move 1/3 of the way from its prevailing level at the time of the forecast to 65% when the actual employment observation is taken a year later.  Unsurprisingly, given the steady decline in the participation rate, employment has undershot this forecast in every observation in the sample period.   Call this one the "Kocherlakota model", if you will.

*   There was an interesting debate in the comments section of Friday's post about whether the benefits of increasing (generally older) savers' returns outweighed the (possibly ineffective) efforts to try to increase youth employment.  Further to this, one interesting dynamic that has emerged over the last decade or so is a sharp increase in the labor force participation rate amongst over-65's...which has occurred at the same time as a decline in the participation rate amongst the age 25-34 cohort.   Although it is possible that each of these trends is voluntary, it seems rather unlikely.   Sure, Warren Buffet is having a ball and works because he wants to, but it seems dubious that retirement-age Wal-Mart greeters get quite as much out of their "careers."  It's hard to say for certain that the older cohort is "hogging the jobs" that would normally go to the younger crowd, but it does seem likely that the dearth of fixed income returns on offer has exacerbated the issue.

*  Macro Man has talked a lot about "spare capacity" but can never recall sharing one of his favorite charts before.  The trend in US industrial capacity growth rose at a very steady trend between 1967 and 1994 before exploding at the same time as the Internet and telecoms investment bubbles of the late 1990's.   Capacity has remained above the prior underlying trend ever since, which is one reason why utilization rates have trended lower, even during economic peaks.

*  Obviously, more spare capacity means less inflation pressure (and less wage growth!)  The chart below compares the level of capacity relative to the 1967-94 trend with inflation in the durable goods PCE deflator.  As you can see, the relationship looks to be quite good.   Now envisage, if you will, what the last two charts must look like on a global basis,  what the implications are for inflation, and what central banks can (or, more to the point, cannot) do about it.

Now let's see where Macro Man's bike (and mind) take him today....