James Pethokoukis cites, without providing a link, a recent claim by Goldman Sachs economists.
If our ballpark estimate of a 0.7pp understatement of annual GDP growth is close to the mark, the measured growth pace over the past five years of 2.2% might correspond to a true growth pace of almost 3%. This would be closer to the pace one would expect given the pace of improvement in most labor market indicators over the past five years.
They think that inflation is overstated, and therefore real output and productivity are understated. The official estimates face several well-known problems, including:
1. For some products and services, quality change is difficult to measure. The statisticians’ estimates of quality improvement tend to be biased in the direction of zero. I’ll bet, for example, that if you were to look into medical care data in the GDP accounts in detail, you would find a lot instances in which quality improvements are mis-classified as cost increases, and almost no cases of the opposite.
2. The process of aggregation is problematic, because people shift their consumption in response to price changes. If college tuition goes up and people switch to using YouTube, it is not clear how to measure the change in real consumption of education services.
3. The nature of inputs changes. To any business, the differences in skills and characteristics of workers matter. In the aggregate productivity statistics, an hour of labor is an hour of labor. There are similar problems with aggregating capital and aggregating output.
You know I don’t like anything that smacks of treating the economy as a homogeneous GDP factory. But if you do think in those terms, then I am inclined to agree with the Goldman folks. The aggregate measures show the GDP factory becoming more and more profitable, even with rather slow growth in revenue (nominal GDP). To get that, you need productivity growth to be better than wage growth. And, yes, that means that an implication of the Goldman view (and mine) is that the people who complain that labor is not getting its fair share would have more fuel to feed their complaints.
It means that real interest rates have been higher than they appeared to be. So does that mean that the zero bound was more binding than we thought? But we got higher growth than we thought??? I think that, on balance, this is bad for people who harp on the zero bound, but my biases have always been against that viewpoint.
Meanwhile, righties, it also means that people who think that there must be inflation going on somewhere, because the Fed is such a wicked institution, are not correct. I expect to get a lot of push-back on that. For those of you who want to use asset prices as your measure of Fed profligacy, you may have a point, but that is a different story as far as I am concerned. For those of you who are new to this blog, I subscribe to a “consensual hallucination” theory of low to moderate inflation. That is, people just operate as if prices are going to keep doing what they have been doing. On the other hand, I subscribe to the fiscal theory of hyperinflation. You get big hyperinflation when, and only when, the government is deficit-spending so much that it can no longer finance its deficits with low-cost borrowing.
Seems like that conclusion is a big problem for monetarists. From their position lower inflation means (or strongly implies) that monetary policy has been tight- and in this case even tighter than they had previously supposed- decent growth in such an environment should be difficult (at best).
I think Goldman is off the mark. They believe that productivity growth over the last 5 years has been off by 0.2%/yr, or about 1% of GDP, due to understating microprocessor/computer performance improvements alone? It’s not like there isn’t already quality adjustment going on. That’s $170 billion in a $17 trillion economy. For context, in 2014, the entire microprocessor industry only brought in $67 billion in revenue.
I have the same skepticism for free content. 0.5%/yr for 5 years is $425 billion of missed growth in a $17 trillion economy. Is it really understated to that degree?
Combining the two, we’re talking $595 billion, or nearly $5,000 per household. Is the typical household really better off by $5,000/yr from these missed intangibles? Global trade revenue for music sales is $15 billion. Total cable industry revenue is a bit over $100 billion. It just seems ridiculous to me that, literally in the last 5 years, the incremental gains of this free content industry and understating microprocessor improvements is equivalent to recreating the cable/music industries 5x over. It’s more than the value of new car sales in the US!
I just can’t care for two reasons I can think of. First, If it doesn’t boost revenues than why should anyone really care? Second, technology progresses. What I care about is how policy is hurting technological progress, not how it mismeasures it.
Do they mention the unprecedented and possibly historic and permanent drop in labor participation? I am tempted to agree with Metallica when they say “nothing else matters.”
I am sympathetic to the inflation-is-overstated idea, but I don’t see it justifying the conclusion about “the pace one would expect.” This pace is calibrated by historical trends which also overstate inflation. To make the case, one needs an argument about why inflation is now being misstated to a greater degree than in the past. It will be very difficult to find sufficient *new* sources of mis-measurement to close the gap.
This all sounds like hand-waving in an attempt to salvage the failed experiment in monetary policy.