Productivity Bad, Economy Bad

by: Steven Hansen

Pundits seem to comment on elements outside their comfort zone (areas of real expertise). Everyone is guilty from time-to-time. I know my comfort zones, and one is in labor and labor dynamics which formerly was the major part of my last life.

My least read analysis are ones on productivity - a particularly boring subject which most understand is important to profitability and competitiveness. Some believe it is a major building block of GDP. (The way productivity is calculated by the economists - it uses GDP components.) Following economic updates on productivity are the equivalent of being forced to watch snail racing on ESPN.

The graph below is the bean-counters and general economic view of productivity.

Productivity in the above graph is calculated by dividing real output by hours worked. If a component was outsourced, this would be considered a productivity improvement. In my posts, I have constantly complained about this view of productivity:

My view of productivity is one of an industrial engineer, while the Bureau of Labor Statistics (BLS) are bean counters using a simple hours vs. output approach. Although one could argue that productivity improvement must be cost effective, it is not true that all cost improvement are productivity improvements.

There is no way for a bean-counter to understand productivity. They only understand that a product is being made cheaper, and not exactly why. But not understanding underlying dynamics sets you up for drawing the wrong conclusion. From my point of view, real productivity has improved little in the USA since 2000, and this is part of the dynamic which led into the Great Recession.

SF Fed Macroeconomic Researcher John Fernald comes closer to quantifying real productivity (his paper at the SF Fed). He utilizes a multifactor method of quantifying productivity.

Multifactor productivity combines estimates of labor inputs with estimates of other factors of production, such as capital, in proportion to their importance in production.

The chart above identifies four distinct periods of productivity growth.

  1. Ten years of productivity slowdown beginning in 1973.
  2. A period of modest growth of productivity that continued into the mid-1990s.
  3. A multifactor productivity growth spurt of 1.7% per year.
  4. A dramatic decline beginning in 2004 in the rate of growth of multifactor productivity to about 0.5% per year caused by labor and capital being underutilized following the initial recovery from 2001.

The Chicago Fed has mused over this multifactor productivity slowdown beginning in 2004:

    • Is it perhaps simply a problem of measurement related to the increasing share of the economy devoted to services-in particular, business and financial services-for which it is difficult to measure output (and, hence, productivity)?
    • Or is it perhaps due to a more widespread problem with the measurement of intangible investments?
    • Alternatively, might it be due to the exhaustion of the gains from the information technology revolution?
    • Or to declines in the quality of education and, hence, the quality of the labor force?
    • Or even to declines in government investments in infrastructure?
    • Depending on the answer, slow measured productivity growth may be consistent with continued rising living standards or a period of stagnation in the developed world.

I can only shoot holes in this multifactor productivity methodology without offering an alternative. The economy is too complex to perform a real productivity estimate as each portion of each sector must be modeled using similar baselines. So one must step back and look for a proxy which provides a "good" (but imperfect) answer. An old boss, after smashing me on my head with my slide rule after making an order of magnitude error told me:

If you do not know the answer before you calculate, chances are your answer will be wrong.

Calculations are done to prove your answer - and if the answer and calculation results differ, you need to go back and re-postulate and / or recalculate.

This multifactor productivity approach is an improvement as it calculates productivity closer to my answer. Productivity growth historically is never a straight line as the BLS productivity suggests. You cannot have real productivity growth without investment (tangible or intangible - or sweat in the case of a small business) - and this is why the multifactor method provides a better answer.

My belief is that

  • the consumer demand curve began to change somewhere around 2000 creating too much USA capacity slack (you do not invest when you have excess capacity);
  • in the same period the free trade agreements started to take hold moving investment overseas.

Even saying this is misleading - productivity is not a command feature of the economy. It follows in the aftermath of a significant technological breakthrough - say the personal computer. Adding to misery, productivity improvement implies less labor to do the same task - how much productivity improvement do you want with employment slack.

Productivity is one of the economic headwinds.

My weekly economic review is in my instablog where I headline with the very good personal consumption and income data from Friday. I wish everyone Happy Holidays / New Year.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.