What role, if any, does secular stagnation play in the flat growth of wages since the recovery?

Question

Many high profile economists and academics seem to be wrestling with this issue, but from all I've read, I still don't have a clear sense that there is a common consensus. I'm referencing Tyler Cowen, Stiglitz, Callum Jones at NYU and others.

Answer 1

The question seems to assume that secular stagnation leads to sluggish wages. I would say that sluggish wages is leading to secular stagnation.
So what is causing sluggish wages? Monopsony and govt policies to maximize after-tax corporate profits while lowering minimum wages and worker's ability to seek better wages.

Also firms no longer set wages according to the marginal value of labor. Now firms determine a price point for their products and services that will maximize profits. This process keeps wages low in order to support the price point in the market.

Also, with interest rates low for so long, inefficient firms are able to stay in business. They are not forced to be more efficient. These low-efficiency firms need to pay lower wages. The result is that more efficient firms are able to pay lower wages even when they are able to pay higher wages.

Comments

I'm visualizing the loss of "net social benefit" in terms of utility and dead-weight loss to the economy. I agree with both your points about the causes of downward pressure on wages, namely that:

1. Low efficiency firms are able to stay in the market for longer, and
2. Monopsony (I'm thinking about Amazon) allows for a narrower employer market, or fewer firms which have to compete for the same pool of labor.

Over the mid to long term I don't see how this is easily corrected for. Will we reach a point where efficiency can no longer be increased and reach an "efficiency equilibrium" where marginal output cannot be increased by *human* productivity?

---

I can't be really sure about that because the workers that are going to join the firm monopolizing the market won't be the same number. A centralized company tend to have centralized own services and this is reflected in a necessity of less workers for everything. For example: cleaning, marketing or security. I have seen that type of firm's centralization in Spain with Mercadona case and a thousand of firms with their independent offices, independent warehouses, etc, vanish from the market and the centralized firm need far less workers to run their operations than the other firms. It's a kind of company model as Amazon. Amazon needs less workers than thousands of normal stores. I think the real problem here is the companies with a good revenue is paying a bad salary to their workers because the trend as you say is widely accepted and or you choose that or you need to find another inefficient firm with a lower wage per month. In my opinion encourage business competence between firms is a good reason to give chances to more inefficient firms but if always the labor market is divided in efficient firms wages and inefficient firm wages. I think that the better way to make those inefficient firms vanish or improve like you suggest is little by little. It's totally correct that a optimal economy is working fast and with good skills and you can't reach a well developed economy running bad ideas or bad production systems. Thank you for your explanation Mister Lambert.

---

Derrick, I see the "efficiency equilibrium" as an effective demand limit based on labor's power to consume. For me, a recession is supposed to clean out inefficient firms so that productivity can once again increase. My analysis of the data after the recent recession shows that inefficient firms were saved by banks probably to keep jobs from being lost. But the jobs should have been lost in order to increase efficiency and productivity, which we know did not increase.

---

Recent models of labor markets with (IMO realistic) search and matching frictions don't have a generic w = mpl result. Take the baseline DMP model, for instance. As long as both firms and workers have non-zero matching surplus, any wage is valid, ie. by throwing out the assumption that labor markets clear, wages aren't determined by that condition. Wage indeterminacy in these models is usually, though not always, resolved via a bargaining problem which allocates the total matching surplus. Thus, the share of surplus going to a worker may be substantially below their mpl if they have low bargaining power and a lousy outside option. Although I haven't seen a model with truly micro-founded bargaining weights (and truthfully, I haven't been looking closely), one would imagine that in an economy with chronically low labor market tightness and monopsony power, the labor share of this surplus could be stunted for quite some time.

---

Good insights ejfeilich. One economist who has done work on ideas like micro-founded bargaining weights is Samuel Bowles. His book and some online material shows his work. Here are some possible links for you.
http://tuvalu.santafe.edu/~bowles/WealthInequality1999.pdf
https://pdfs.semanticscholar.org/6378/2dfbb7c6148ee79a33370fa1a5e990e149b8.pdf
you can do a google search for "samuel bowles Microeconomics - DRAFT - Master HDFS pdf"

Answer 2

I don't think we need 'secular stagnation' to address this issue because it means different things to different people but I think that this touches on those themes.  One extremely important stylized fact of the last 10 years is the extremely sluggish growth rate of productivity in the US.   The following are 10-year average labor productivity growth rates compared to prior 60 years.  Nonfinancial corporate 1.1% v 2.3%.  Nonfram business 1.2% v 2.3%  Real GDP per person employed 1.0% v 1.8  Manufacturing 0.7% v 3.6% (this only overs 1988-2008 because of data)  Total economy from BLS data 1.0% v 2.1%.  Thus we have expenditure, income, and output based measures and all show a very slow growth rate of productivity over the last 10 years of about 1% versus a long-run average of around 2%.  Much discussion of wage growth--especially in the media and political circles--has made scant reference to the slowdown in productivity.  The question to be answered, in my view, is why has productivity growth slowed so much.  Here a simple Cobb-Douglas type model and aggregate data on the capital stock and hours worked gives us some insight.  If Y=AK^bL^(1-b) then Y/L=A(K/L)^b (I am suppressing the trend for simplicity but this changes nothing) and taking logs gives ln(Y/L) = ln(A) + bln(K/L) and taking differences reduces this to (Y/L)% = b(K/L)% where b is the capital share of output where % equals percent change.  Take differences again and you have the change in productivity growth equals the capital share times the change in the growth of the capital labor ratio.  In the TFP data K/L grew at 2.6% from 1948-2008 and has fallen 0.2% per year on average from 2009-2017.  So K?L is growing about 2.8% points per year more slowly than the long run trend.  If we take the capital share to be about 1/3 then this would account for almost all the productivity slowdown.  The question is why the slow growth in the capital stock relative to labor in this expansion.  Perhaps this is where secular stagnation comes in.  My view is that it has been in part due to the rising US corporate tax rate relative to the OECD average and the cut in the corporate tax rate last December will boost K/L over time and raise productivity growth.  So, by the way, will the demographically induced slowdown in the L to about 0.5% per year from around 1.5% per year.

Comments

John, your data points are instructive.  2009-2017 were the years of an unacceptably slow "recovery" from the Great Recession.
(which it seems starting in 2016 became a bit more robust).   The slow growth in aggregate demand (in part caused by the "recovery" for the top 10% of the population -- and not much of one for the rest of the population) was probably the main reason for the cutback in productive investment.

First of all, investment had fallen to a dramatically low 12.7% of GDP at the depth of the Great Recession.   However, unlike previous deep recessions (1982-83 and 74-75 for example) when the Investment to GDP ratio rebounded quickly (reaching 17.2% in 1976 and a dramatic 20.3% in 1984), in the years after 2009, investment as percentage of GDP rose much more slowly.  It took three years from 2010 through 2012 for investment as a percentage of GDP to reach 15.5 percent.   It reached a maximum of 17.1 percent in 2015 before falling slightly in 2016 (it remained below 17% for 2017 as well).  The percentage achieved in 2016 was about the same percentage that had occurred at the depth of the previous recession in 2001.

Meanwhile, the financial sector recovered dramatically (with a lot of help from us taxpayers and the FED) and so the financialization of the economy (which many who support the secular stagnation analysis believe is a result not a cause of stagnation) proceeded apace.

I am very skeptical that there was any significance to any (relative) increase in the corporate tax rate during this period --- the rest of the OECD did a much worse job recovering from the Great Recession (there was a double-dips in Britain for example).

To follow up on the secular stagnation analysis, one has to go deeper than Larry Summers' speech to the IMF and subsequent interviews.   The best sources are from the journal Monthly Review and the various books written in the tradition of Paul Baran and Paul Sweezy's MONOPOLY CAPITAL.   A good introduction would be John B. Foster and Fred Magdoff, THE GREAT FINANCIAL CRISIS (Monthly Review Press, 2009)

---

I'm sorry sir. I have a Cobb-Douglas graph in a Macroeconomic book but the data is uncompleted and I see a rising trend on productivity after 2010 similar to former rises. When you are talking about the productivity rises. Are they global, taking into account the high decrease on productivity between 2008 and 2010 (what can cause the global datum to decrease) or is it a continuous datum with not considering the decrease of the Great Recession? It could change the results in my opinion. I don't find any data on Internet. Thank you in advance.

Answer 3

I agree with both of them. The low wages trend is crossing continents due to two facts in my opinion. 1. Is the success of economies like the Chinese. The companies producing in this kind of lowcost economies are earning a great rate of return of investment and a great profit and the companies around them want to earn similar rates of return of investment so a way to do that is to lower their own wages. This create a trend of low labor cost searching. Besides, this model of economy tend to absorb wages from other countries because of their low level of human cost and the bad working conditions of their workers. When I say that I refer to the fact that the Chinese economy has a big level of domestic consumption from abroad companies paying wages in their economy (I'm not going to take into account their huge corporative debt). I live in Spain and after the crisis the government started talking about austerity, it's a common word nowadays in countries with economical problems. The workday was lenghten to 9 hours and the minimun wage stayed at 600 euros. The effect of this is a slow growth of the economy because the demand is too low. In a first step of the recovery it's normal to have low wages because the economy have to repair the damage and the companies need to fix their balance sheets. But in the long run, to mantain these low wages is reflected in an unemployment rate of 15.1% ten years after the crisis. In the US this trend is not so tough and wages are higher but a low level of salaries can create a poor demand and so a low growth.
2.  The maximization of profit. Taking into account that the wages are one of the biggest expenditures of a company, the "greedy" share-holders are trying to minimize the labor expenditures. A machine is buyed and the cost of it is reduced as the time goes (taking into account amortization) but the wages cost is paid every month again and again and a good way to maximize profit is to mantain a low level of wages and a low number of them. This is bad for the economy because a low demand again create a low growth and a a low life standard (taking into account that the more the economy grows the faster it collapses).
Talking about productivity, it's normal that the investment grew when the credit started to flow again and when you have invested in capital it's normal that in a few years that investment decrease because everybody has new machines and they can wait some years before buying new ones. So it's normal that the last years the investment fell and this create a slowdown in the economy because the firms working for other firms see their revenue diminished so then they have to lower wages what create a greater slowdown in the economic activity. The slowdown on productivity can be caused by taxes like Jonh Ryding indicated or maybe because the mechanization or the use of new tecnologies had more weight on productivity rises than the actual developments. Anyway I think that wages are not in accordance with productivity levels. In my opinion this trend I explained before is not allowing a level of wages in accordance with the marginal desutility of labor, and the workers were in a bad position after the crisis to bargain better wages. Fact that is still happening years after. Everybody wants a job and the marginal utility of leisure is very low.
I'm sorry for the editing. I hope don't mess up your mail. My English is not perfect. Greetings.

Answer 4

I think Ed Lambert is correct. "Secular stagnation" is probably not the best label for the worry. The worry is that financial markets have gotten themselves wedged into a situation in which frequently and for sustained periods of time it is the case that the full-employment real safe short-term Wicksellian neutral rate of interest turns out to be less than the negative of the central bank's inflation target. In a flexible-price full-employment economy, the economy deals with this and maintains full employment by having the price level drop instantaneously and discretely whenever this occurs in order to generate the extra inflation needed to get the market rate at the zero lower bound to its value needed for full employment, the real safe short-term Wicksellian neutral rate of interest. This was one of the major (but I think often overlooked) points of Krugman (1998) <https://www.brookings.edu/wp-content/uploads/1998/06/1998b_bpea_krugman_dominquez_rogoff.pdf>

But in a sticky-price economy, things get messy when frequently and for sustained periods of time it is the case that the full-employment real safe short-term Wicksellian neutral rate of interest turns out to be less than the negative of the central bank's inflation target...

That worry is correlated with low expected productivity growth, which is correlated with low real wage growth. But they are not the same thing...

Yours,

Brad DeLong