The make-or-buy decision with respect to human capital

From a National Academy of Sciences report:

As long-term employment becomes less common, new ways of providing for health care and pensions for all workers need to be considered that transcend their relationships with particular employers. For example, one option would be to institute portable pension plans administered by membership organizations dedicated to the well-being of their members.

Pointer from Timothy Taylor, who includes this quote from the report:

Many employers are increasingly viewing their relationship with employees as a short-term commitment rather than a lifelong investment. As Manpower Group CEO Jonas Prising recently put it, `Employers have gone from being builders of talent to consumers of work.’

Note, however, that this supposed decrease in the share of long-term relationships between employers and employees is not so evident in the data.

Still, I think that the trend is toward firms making more and more use of generic software. (I predicted this almost twenty years ago.) In fact, software-development specialists make more and more use of generic software.

Gary Becker developed the distinction between specific human capital, which is tied to a particular firm, and generic human capital, which can be used anywhere. The more that firms outsource non-core functions and make use of generic software the less they need to invest in specific human capital. Thus, they will tend to do less talent-building and act more like “consumers of work.”

This entry was posted in business economics, links to my essays, Timothy Taylor is my Favorite Blogger. Bookmark the permalink.

10 Responses to The make-or-buy decision with respect to human capital

  1. Scott Gustafson says:

    Go look at the JOLTS data, specifically total separations on Chart 3. About a third of the labor force changes employers every year. Given that, we should not be getting either retirement nor health insurance from an employer – not if we value long term planning and stability in those aspects of our personal lives.

    Employers must also question investing in employees that don’t stick around.

  2. Kevin Dick says:

    Note that this hypothesis conflicts with Garret Jones worker hypothesis. Organizational capital is, pretty much by definition, distinctive so producing it will require a fair amount of specificity.

    There are two obvious alternatives. Firms outsource non specific processes (which in turn are specific for the firms that execute them). And/or, there is a stratification of workers into generic execution and specific organization capital creation roles.

    To me, the former seems like a more attractive option as it’s new patterns of specialization all the way down. Of course, reality has this annoying habit of conflicting with my aesthetic sense.

  3. Lord says:

    Investment in human capital may shorten rather than lengthen employment through promotion, transfer, and advancement, while consumption of labor may lengthen it by lowering advancement and dynamism.

    • Dan King says:

      Right. So the “generic software” hypothesis really means that there will be less specialization, and therefore also less trade (or vice versa).

      It’s a depressing scenario. Fortunately I don’t think it’s true. Entrepreneurs will always push the edge to find new opportunities. There’s nothing generic about the edge.

  4. collin says:

    Note, however, that this supposed decrease in the share of long-term relationships between employers and employees is not so evident in the data.

    In my office this is very true and we have not had a major layoff in 6 years whereas it was normal September/October activity. I think the issue is our office cut so much in 2010 we no longer have a talent bench to speak of. Also, in our business, human resources hates that our talent does not have any loyalty any more.

    1) The 401k and health insurance employer based systems need to go away. This was the primary reason why I supported Obamacare and I am surprised the health care employee system would weaken after 2009/2010.

    2) I noticed it is liberals not conservatives discussing personal pension programs. The 401K system is consumer nightmare and I have three accounts which a fuckin’ pain to deal with.

    3) For economic liberatarians this is the chief reason why nobody thinks of loyalty for community. My conservative father-in-law says people just don’t care and I said we are following the lead of our economic leaders like Koch Brothers and Jeff Immelt.

    4) And finally, isn’t this reason too many people go to college? They know their employers will have no loyalty so it is better to develop skills and resumes.

  5. JK Brown says:

    We’ve an example of specialized vs generic employees with the industry unionized labor. The generic labor, though skilled, could easily be utilized by various employers. The union took over administration of healthcare and retirement programs independent of specific employer. The unions also provided training programs that were cross-company compatible. Of course, the pros and cons of union labor have been well documented.

    On the other side, with James Bessen (May 20160 describes the move from employer specific human capital to “generic” workers, due to machine standardization, as the textile industry developed. There the workers gained in bargaining power as they could go down the road for work.

    In an EEVblog interview with Jack Ganssle, an embedded expert, they discuss how many companies now do not spend the time optimizing a chipset, but rather throw in a generic chip/memory running Linux and hire “dollar an hour” programmers to write the software. This is a major source of the IOT zombie bot internet problem as those unpatched and often wide open Linux systems spread everywhere. Even beyond that, it is often smarter to pipeline several packages from other vendors rather than write/maintain an integrated app.

  6. Bryan Willman says:

    Some of this strikes me as so obvious as to be flummoxing – this is news?

    Decades ago, places like Microsoft outsourced any “generic” task (receptionist, janitorial services, many real estate functions, part of the legal team, retirement program admin, the list goes on and on.) I just assumed everybody else did as well.

    And many have observed that the real wealth and power in MSFT or Amazon or Google flows to the groups and jobs that are the company specific revenue makers. Ad words for google. Windows or Office or perhaps now Azure for MSFT. Ordering or pricing or aws for amazon. Basically everything else risks being outsourced, because it’s “generic” in the sense of “not being the core of the revenue flow”

  7. B. Reynolds says:

    I get my auto, home, and life insurance outside of my employer. Somehow we don’t need a government cluster-fudge like Obamacare or a National Health Service to make that happen in those markets.

    This awful tie between employers and health insurance has been a well-known issue for decades. It says a lot about their lack of enthusiasm for markets that Republicans never lifted a finger to correct this unfortunate disaster of history.

    It would have been an ideal issue for the Neoliberals to champion 25 years ago.

  8. Thales says:

    “I get my auto, home, and life insurance outside of my employer. Somehow we don’t need a government cluster-fudge like Obamacare or a National Health Service to make that happen in those markets.”

    Other than the fact that it’s currently not permitted by law, I’ve always wondered why one can’t just buy a whole life medical policy the way one can for life insurance. It would have the advantage of not only encouraging lifetime personal responsibility, but twenty-somethings could get excellent life-time rates and benefits for if/when they contract that chronic condition in middle age or beyond.

    • sort_of_knowledgable says:

      One, if employer subsidized health insurance is usually available then personal whole life medical would not be useful.

      Two, insurance companies can hedge their costs for whole life insurance because they can price the insurance based on the 30 year bond rate. But they don’t have anything equivalent for medical costs 30 years out so they would have to price it so high that it is not worth while.

Comments are closed.