Captial Bias, Robots, Employment and Big Data

For years I’ve been writing about my belief that at some point technological progress will impact the economy enough to make us question our assumptions about growth, wealth re-distribution, and welfare.  And finally, based on a recent smattering of articles, it’s no longer a topic found only in Sci-Fi novels.

The talk at this point is mainly about admitting there is a problem.  The final paragraphs of these essays generally state things like “this brings up some uncomfortable questions”.  The assumption is that capital bias will lead to the continued decline of median wages which necessarily leads to the decline of the median standard of living.  My beef is twofold.  First, in a Democracy, you would expect the people to quickly realize that they have less money and vote for increased wealth redistribution (note, this is an apolitical statement).  Second, in the graph below, notice how GDP has continued to increase while median household income has stagnated?  The assumption by some is that we can’t afford the Keynesian prescription of spending our way out of the “jobless recession”.  That is true only if your assumptions about Government revenue map to tax rates on median income.  Until about 1983 that was a good assumption.

The seemingly obvious solution is to re-link tax revenues to GDP, which continues its inexorable rise. If this can be done any questions about deficits seem almost pointless.

Here is one more graph which pretty clearly shows that something is increasing the duration of



unemployment over the past three recessions, interesting that there is no similar trend before 1983 given the chart above.  Also, note that in every recession since 1981 duration of unemployment has grown.  That might be a complete coincidence but it sure looks a lot like something changed in the early ’80s which is also roughly the same time that computers were introduced into the workforce en masse.

And finally, my ankle has been gnawed by the big data bug.  I’m currently using R to run statistical analysis on some fairly large data sets and I’m think that I might be able to make my points better with some animated charts…  Maybe I’ll do it in my 2013 post, assuming the robots haven’t taken over…

Note – Some of my related posts are many years old and don’t reflect my current take on economics and/or politics.

5 thoughts on “Captial Bias, Robots, Employment and Big Data

  1. Robert Rasmussen

    Hi Kirk,

    Nice blog! A couple of thoughts:

    Not sure that the real impact of computers in the workplace took shape until the early 90′s not 80′s. This really took shape with networked PC’s and technologies like Banyon, Netware and Windows 3.11

    What I recall did shape things in the 80′s was the rise of Japanese manufacturing power (be it Toyota or Casio) on the tail end of the oil shortages of the 1970′s and supply-side economics of Reagan which had a positive effect on employment also ushered in a new era of globalization, end of the cold war, and a globally supply-side economy. I would guess that absent the technology revolution of the 90′s and our leadership in the advent of the Internet and e-commerce, the US would have really been in trouble by now?

    Perhaps Americans have enjoyed an unsustainable quality of living and globalization will be our eventual downfall. The 12 percent of the world’s population that lives in North America and Western Europe accounts for 60 percent of private consumption spending, while the one-third living in South Asia and sub-Saharan Africa accounts for only 3.2 percent.

    So perhaps on a macro-economic basis, any progress, be it spread of capitalism, or technology will have short term benefits for the US, but long term serve to re-distribute consumption globally. On a micro-economic basis I think the short term benefit in the US will be to less and less of an extent any relief for the working class, but only serve the pockets of those with ownership stakes.

    Good news for the very poorest and very richest in the world, but the middle class will continue to shrink and that is where unemployment rates reflect reality.

  2. Kirk Post author

    Thanks for the comment Robert. Agree that some portion of the falling wage trend is due to globalization but I disagree that in a world of growing GDP (see chart above) the standard of living necessarily has to fall in the West due to global wage arbitrage. Much of the recent talk of onshoring previously offshored labor is a result of robots having lower wages than even 3rd world laborers, and the shipping is free!

    So maybe a rising tide really does lift all boats, but only if we resolve the holes in the middle class hulls with modern tax policy. I just wonder how you avoid a race to the bottom in terms of taxes which is currently happening with US state tax incentives (see Texas).

  3. Greg R Perry


    Do your figures for compensation include health care costs? I’ve heard a lot of anecdotal evidence that total compensation has actually been keeping track with growth in productivity and GDP, but employees haven’t seen it in their paychecks because employers are diverting it to pay the rising cost of insurance. These overhead commitments could also explain employer reluctance to hire after a downturn. Maybe the technology boom occurred in the medical industry.

  4. Kirk

    Good point Greg, but looking at that chart, assuming wages did keep up with GDP and productivity when you factor in health care, that would necessitate that half of our wages are comprised of health care benefits. Only about 160 million workers receive health benefits and the $ value of that benefit averages less that $5k per year. So if you factor in health benefits as wages (assuming it’s not already) it might have a small effect on the numbers, but not enough to re-align wages with GDP.

  5. Kirk Post author

    Greg – Quick followup, NYTimes article this morning addresses your point specifically:

    Some economists say it is wrong to look at just wages because other aspects of employee compensation, notably health costs, have risen. But overall employee compensation — including health and retirement benefits — has also slipped badly, falling to its lowest share of national income in more than 50 years while corporate profits have climbed to their highest share over that time.


Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>