Notes on the Solow Residual and Aggregate Supply

A recent post from the Institute for New Economic Thinking, entitled "The New Normal: Demand, Secular Stagnation and the Vanishing Middle Class", deserves more responses from other bloggers than it is likely to receive in a time of political volatility. First, a point of agreement, with Servaas Storm's assessment that the U.S. is becoming a dual economy:

...two countries, each with vastly different resources, expectations and potentials...

Nevertheless, I have little choice but to pour "cold water" on his hopes that a divided whole can somehow be reunited, via what would essentially be the same "dressed up" equilibrium (more income for all) model - albeit with a drastic retake on monetary policy. Fortunately, that does not mean there are no good alternatives. In a way, I can understand why some would hope that modern monetary theory could address both budgets and income divides, when presently, precious few other options are on offer. Just the same, I'm afraid this approach would only disappoint, were it ever given a serious chance in Washington.

One interesting aspect of Storm's post, was the suggestion that the Solow residual could simply be disregarded in the future. He feels it serves little purpose - even to the extent of standing in the way of long term growth:

The task looks Herculean because, as most economists would argue, the U.S. is riding a slow-moving turtle and there is little politicians can do about it. This view is founded on the evidence of a secular decline in aggregate total-factor-productivity (TFP) growth - a widely used indicator of technological progress, fondly known as and measured by the "Solow residual". Dwindling TFP growth, which is in this view taken to reflect a general malaise in exogenous "technology-push" innovation, reduces the rate of growth of potential U.S. output... 

The U.S. is suffering from two interrelated diseases: the secular stagnation of its potential growth, and the polarization of jobs and incomes. The two disorders have a common root in the demand shortfall, originating from the "unbalanced growth" between technologically "dynamic" and "stagnant" sectors, which - crucially - is bringing down potential growth. To understand how the short-run demand shortfall carries over into the long run, we must first rethink the Solow residual, which economics textbooks define as the best available measure of the underlying pace of exogenous innovation...But it can be shown, using national-income accounting, that there is no such thing as a Solow residual, because it must equal - as a matter of accounting identity - either "weighted-factor-payments" growth or "weighted-factor productivities growth".

First: Those who read his post in its entirety, will note that Servaas Storm also has a different perception of what William Baumol's equilibrium imbalance implies, than what I recently stressed. As far as I know - and someone please correct me if I'm wrong -  Baumol never suggested that wage equalization in stagnant sectors would make them more "dynamic". After all, wage equalization in non dynamic sectors would mean both higher wages and higher costs as well. When Baumol said time based services could eventually impede growth, it was because income gains in these sectors translate into increased costs for others, displacing more efficient dynamic sectors.

However, real gains come from making income go "much further" via innovation in basic equilibrium settings (consumption gains), rather than trying to "stretch" income levels to fit a given equilibrium pattern. While equilibrium imbalance may appear as an income problem, the underlying problem is an imbalance in resource utilization, or in terms of the input (investment for economic access) that is now required for the output of time based product.

At the very least, present levels of indeterminate output are being held in check, relative to the wealth creation that is determinate output. Why should this matter? We are still adding to the input requirements of time based product, even as we gradually come to expect less output from the actual process. Indeterminate time product output, translates into more economic uncertainty. Why save every extra dollar in one's working years, to wait and travel in retirement, if doctor's bills only end up requiring retirement savings? Consequently: In aggregate, we keep trying to pay for time based services with time we don't actually have, which leaves everyone perpetually behind the starting line of opportunity, with a negative Wicksellian interest rate.

If we were to deny the Solow residual and turn modern monetary theory into fiscal "monetary" policy, this cost/input process would be exacerbated - both by the loss of measure for tradable sector activity, and the extent to which automation does affect non tradable sector services generation. Accounting identities do not recognize the validity or differential of aggregate input, in relation to aggregate output. As time based product becomes more prevalent in relation to tangible goods, product indeterminacy means taking the chance that more individuals will become inclined to disregard the value of knowledge based product. Presently, aggregate time value does not have a reliable relationship, with knowledge use and other forms of resource capacity. Discarding the Solow residual concept would only make this situation worse.

While we have greatly benefited from indeterminate knowledge based product in recent centuries, taxpayers are only inclined to tolerate service product indeterminacy, up to a point. One only hopes that we will not abuse the trust of today's knowledge use patterns which fiat monetary policy makes possible, via spontaneous coordination at a national level. Few taxpayers question the output gains of tradable sector output, because this determinate output is an obvious indicator of progress. Whereas even though non tradable sector time based product holds tremendous value, all along we've have to take the vital connections between labour hours and aggregate supply, on faith.

What can be done? Instead of doubting the integrity of the Solow residual - and traditional monetary theory for that matter - create real space for new long term growth. Generate time based services growth, as a form of determinate output. While aggregate time value would remain relatively constant, ongoing management in knowledge use, would mean an entirely new way to capture productivity gains.

Today's indeterminate services product may stand a better chance of retaining its nominal income value, with the addition of time based services creation as a determinate form of output. By creating space for services that can be readily be measured, everyone's time would gain additional value in the marketplace. One way to think about the process is that determinate time based service product, or time arbitrage, would be measured differently than the Solow residual, because of time constancy in employment for time based product. However, it is the new ability to measure service creation at the margin, which ultimately helps to preserve the measurement integrity of the Solow residual, for both tradable sector production, and the technology gains of today's indeterminate services product as well.

Fiat monetary formation, thus far, has allowed indeterminate services product to generate a secondary marketplace which reflects the wealth creation of today's primary markets. What is particularly important, is that the second group exists in relation to the first, in terms of aggregate resource capacity. One of the main problems with modern monetary theory, is that it would not be able to faithfully represent this relationship, a fact which could put a considerable amount of knowledge based product in jeopardy. Let's hope that doesn't happen.