Friday, May 12, 2017

Dollar Domination, Robot Monetary Overloads, and Closing the AD Gap

Some assorted musings:

1. From this week's podcast with Ethan Ilzetzki comes this amazing figure. It shows that approximately 70% of world GDP is tied to the dollar. The implication is staggering: the FOMC is setting monetary conditions for much of the world.

2.  Greg Ip argues our robot fears are misplaced. If anything, we do not have enough robots destroying jobs:
From Silicon Valley to Davos, pundits have been warning that millions of individuals will be thrown out of work by the rapid advance of automation and artificial intelligence. As economic forecasts go, this idea of a robot apocalypse is certainly chilling. It’s also baffling and misguided. Baffling because it’s starkly at odds with the evidence, and misguided because it completely misses the problem: robots aren’t destroying enough jobs...  
This calls for a change in priorities. Instead of worrying about robots destroying jobs, business leaders need to figure out how to use them more, especially in low-productivity sectors.
Okay, what industries fit this description? Greg mentions the usual suspects: education, healthcare, and leisure and hospitality. Here is another one: central banking. It has not changed much in many decades and probably has low productivity. So maybe the hidden message of Greg's article is we need to get robots running the Fed?  My answer: sure, but only if they are targeting a NGDP futures market as proposed by Scott Sumner. This approach, after all, is fairly automatic by design and therefore conducive to our robot overlords taking over the Fed.

3. The CBO estimates the full-employment level of aggregate demand. They unfortunately call it "potential nominal GDP' which is horrific since the potential is truly infinite--think Zimbabwe's NGDP in 2008--so ignore the official name. The idea behind the measure is reasonable: what level of aggregate nominal spending is consistent with full employment. Here is the series lined up against actual aggregate demand as measured by NGDP. 

The figure indicates there was a sharp collapse in aggregate demand during the crisis and the full-employment level has only slowly converged to it. Therefore, there was both a sharp decline in aggregate demand (a cyclical shock) and a downward adjustment in the full-employment trend level of aggregate demand (a structural shock). This understanding is consistent with the recent Fernald et al. (2017) BPEA paper that found that there was both a demand shortfall (that ended by mid-2016) and a decline in potential real GDP. So the CBO's full-employment level of aggregate demand seems quantitatively reasonable.  

For fun, I plotted the difference between the full employment level and actual aggregate demand--the AD gap--against the latest hip labor market indicator. It is the working-age employment rate (a termed coined by Jordan Weissmann) which is more commonly known as the employment-to-population rate for prime-age workers (25-54 year olds). Nick Bunker has been a tireless advocate for using this measure to replace the unemployment rate as the headline labor market indicator. 

So how does the working-age employment rate measure up against the AD gap? Not too bad according to the figure below. The relationship is not perfect, but it is strong enough to indicate that the continued upward recovery of the working-age employment rate over the past few years has been largely due to cyclical factors. 


  1. I'm also a fan of that statistic.

  2. As per the BIS, roughly %60 of forex volume includes $USD, so %70 of global GDP makes sense. Source:

    Within an economic context, the Federal Reserve is truly the most powerful institution on the planet, currently. Yellen is likely to raise in June so long as the stock market, jobs, and inflation remain stable, which they have been (notice she/the Fed rarely look outside the US.. stupid). Weaker nations better get their ducks in a row soon!

  3. Woah, that last graph fits stunningly well for things that are independently and noisily measured.

    I didn't think it was possible in macroeconomics, to get empirical results that are this good.

  4. Unless that Nominal Potential GDP is calculated from the employment statistic and GDP and you are just extracting the employment statistic signal back out?

  5. AD falls when money flows decline. M*Vt = AD, and AD is a proxy for the subset, N-gDp. Vt falls when savings are bottled up and idled in the payment’s system. I.e., contrary to every modern day economist, commercial banks pay for their new earning assets with new money, not existing deposits, or short-term debt. See "The Economist":

    "Because one person’s spending is another’s income, such saving can drain away demand and lead to recession, unless the funds set aside by savers are reinjected into the economy through lending to those looking to invest: as when banks lend savers’ deposits to businesses, for example."

    The metaphysical error is that banks do not loan out the savings placed with them.