Bigger Stocks, Slower Flows

Nicholas Eberstadt has an excellent summary of recent, largely depressing American economic trends, called “Our Miserable 21st Century.” He begins with an opposition that doesn’t get pointed out often enough- that steadily rising American wealth has coincided with disappointing GDP growth:

eberstadt_figure1eberstadt_figure2

The thing to note is that this is not directly a consequence of income inequality: an increasing income going to the richest in society would still produce an increasing GDP. It could, of course, be an indirect consequence of income inequality. If richer people spend less of their wealth, the rich getting richer can also cause slowed GDP growth. (This is a central aspect of income inequality that Piketty largely ignores in his book.) But it could also be largely distinct from American inequality. If rich people and sovereign wealth funds from around the world are investing in American assets (whether real estate or corporations), this will drive up the valuation of the portion of those assets that American households and non-profits own.

It could also be that the slowdown in GDP growth is a consequence of rapid technological change, rather than a sign of technological stagnation. GDP is a measure of exchange of goods and services, a flow variable rather than a stock variable. Not only has the explosion of the internet pushed a lot of activity into less monetizable forms, but in our ordinary economic lives technological change can slow rather than speed up the process of decisionmaking and contract and agreement that is the heart of the modern world.  The more we base decisions on algorithmic processes that nobody fully understands, the more of a challenge creating consensus around those processes becomes.

This isn’t just a matter of software that nobody understands, though I think people underestimate how much confusion throwing ever more complex analytic tools at simple problems can cause (as Robin Hanson recently said, most things that companies think they want artificial intelligence and machine learning for can probably be better accomplished with decently collected data and ordinary least squares.) My candidate for the biggest driver of the “cost disease” that Scott Alexander wrote recently about (and that I experienced in my own life as the $500,000 school grant for a science lab that didn’t produce a science lab) is the opacity of regulatory frameworks that constrain government action in particular, which are their own kind of incomprehensible algorithms. Why is it taking eight weeks for the state highway authority to resurface a single exit ramp on the highway leading to my job? Only the wise must know, but I bet the wise have also read the state administrative code.

s-and-p-500

Moreover, more economic processes are consolidated within larger and larger firms, which may increase in value (as can be seen from the ever-rising S&P), but intrinsically introduce sclerosis in the flow of goods and services. Take, for example, the dramatic consolidation of finance and shuttering of community banks since the financial crisis: as this Politico article notes, “more than one in five U.S. banks have disappeared—1,708, or more than one every business day—since Dodd-Frank was enacted” in 2010. To a decent approximation, we responded to big banks torching the economy by helping them destroy all the small banks:

community-banks

As Arpit Gupta of NYU observes, this not only affects the finance industry itself but has produced a relative increase in small business borrowing costs relative to larger firms, again pushing economic activity within larger, more opaque and hierarchical organizations:

c4rchzmw8aav_8p

This graph from the Economist, comparing the concentration of industries in 1997 to 2016, makes this point more directly- almost all industries are becoming more concentrated in a few firms:

20160326_woc010

The consolidation of economic activity into ever-larger firms, directed by increasingly opaque software and regulatory frameworks and responsive to global capital flows rather than individual small investors may not be entirely to blame for the divergence between steadily increasing American wealth and slowed American GDP growth, but it sure seems likely to play a role.

 

10 thoughts on “Bigger Stocks, Slower Flows

  1. I agree that “cost disease” and slowing growth are a problem.

    However, don’t think the two graphs at the beginning are relevant to the understanding the problem. It is likely that the net worth graph and GDP graph difference may just be a result of differences in methodology between GDP deflator calculations versus CPI calculations.

    In the “net-worth” graph, that graph comes from dividing nominal net worth by the CPI change. The “real GDP graph” comes from dividing nominal national income by the GDP deflator change. If “nominal net worth” growth outpaces “real GDP” growth, that means people wish to maintain a bigger reserve in their brokerage and bank accounts in relation to their projected expenses. There could be a number of reasons for this — increased life expectancy means you need more retirement savings, more job insecurity, etc. If the nominal numbers are the same, but GDP deflator growth has outpaced CPI growth, you’d really have to dig into insanity that are these calculations in order to know why that is. The difference could be just one of choices in methodology rather than anything real and meaningful.

    Like

  2. You have three different points, at least:
    1. Sophisticated analytical tools cause most industries to consolidate into ever-larger role
    2. Complex regulatory frameworks cause industries to consolidate into ever-larger firms
    3. The consolidation of industries into ever-larger firms slows GDP growth
    2 and 3 seem very plausible to me, 1 I think is more far-fetched. Anyway, my main criticism is that you didn’t really present clear arguments for any of those statements. Yes there are some graphs, but they basically show A and B (and B, and B) happen, rather than A causes B.
    The only tentative argument for point #1 is that “[t]he more we base decisions on algorithmic processes that nobody fully understands, the more of a challenge creating consensus around those processes becomes”, which I think is… OK, but not remotely strong enough to counter the fact that most firms that use more sophisticated tools do so because that arguably makes them more productive (I also think that you slightly misinterpreted that wise post by Robin Hanson in a way that makes it seemingly support your point #1, though I think it does not).
    Point #2 is small a subset of what John Cochrane — who thinks GDP suffers from “death by a thousand cuts” caused by misguided and excessive regulation — argues in his response to Scott Alexander, and it’s not clear why you think that specific mechanism (consolidation of industries) is the main aspect of that macro-explanation.
    Perhaps I’m being too picky, but I do find this post less convincing/solid than your (really high) average, especially when it comes to culture and society.

    Like

  3. I would say that saying that the US is ‘under banked’ is like saying we are ‘under hamburgered.’ We have a surprisingly large number of small banks. Regulated banking is constantly under siege by non bank lending. For example, 30% of new cars are leased and that is done by the manufacturer’s captive finance company. There is unprecedented venture capital, private equity, and business development companies. Loans are originated and securitized and there is less advantage to engaging in that sort of activity in a FDIC insured/regulated bank. Which is consistent with the cost of regulation. But is more due to the regulatory pressure discouraging risk assumption.

    If we are moving from manufacturing employment to service employment, it is harder to grind out capital intensive productivity increases. US manufacturing and agriculture are very productive but that is the result of capital per employee. Is teaching ever going to become more productive?

    Demography can explain some of it. We are dealing with an aging population and low native birth rates. Household formation growth tends to fall directly to the growth curve.

    Since 2008 we have been living with deflationary pressures. Oil and commodity prices collapsed a couple of years ago. I found this shocking as commodity prices were considered an inflation hedge for a number of years. The idea that you can’t print basic materials (like money) was widely believed, yet ‘stuff’ got cheaper. Not only manufactured goods but also materials. But as a whole, we experienced modest inflation. The economic sectors that resisted price deflation tended to be associated with the public sector and high regulation.

    Scale was the single largest and most powerful strategy of the 20th century. Industrialization is a matter of scaling up. The advantages of scale are so important in the earlier stages that even the USSR was considered efficient prior to WW 2.

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s