Ryan Bourne said this morning on Twitter:
Inequality puff pieces and junk papers just can't answer a simple question: how does a larger gap of outturn income reduce productivity growth?
Here are some possible mechanisms.
1. Inequality might demotivate poorer-paid employees because they look to star employees and bosses to help the firm rather than take the initiative themselves. One study of Italian football teams has found that "high pay dispersion has a detrimental impact on team performance." This is consistent with Jeffrey Nielsen's argument that leader-based organizations cause an under-utilization of employees' abilities.
2. Wage inequality reduces (pdf) job satisfaction. And less happy workers tend to be less productive.
3. Inequality reduces trust. As Eric Uslaner and Mitchell Brown have written (pdf):
Economic inequality leads to less trust in two ways. First, high levels of inequality lead to less optimism for the future. Greater pessimism means less trust. Second, where there is a lot of inequality, people in different economic strata will be less likely to have a sense of shared fate.
There's strong evidence that low trust leads to lower growth. One reason for this is that trust helps to overcome "markets for lemons"-type problems of asymmetric information. As Deirdre McCloskey wrote: "business normally depends on a state of trust not on explicit contracts."
4. Where inequality is high, the rich will invest (pdf) a lot in simply protecting their privilege - for example in intellectual property protection, police, security guards, lawyers and so on. This diverts resources away from productive use. Stephen Magee shows (pdf) that there's a hump-shaped relationship between the number of lawyers in a country and economic growth - and the unequal US has too many lawyers.
5. At high levels of inequality, the rich might fear that property rights are insecure - because they might fear revolution, a harshly redistibutive backlash or simply theft. This deters investment. As Ronald Beanbou wrote (pdf): "inequality exacerbates social conflict, which in turn makes property rights less secure and reduces growth."
6. Inequality can prevent a shift to more productive organizational forms. There's reasonable evidence to suggest that worker coops can be at least as productive as their hierarchical counterparts. Which poses the question: why aren't there more of them? One reason lies in inequality. Poor workers lack the access to credit that would allow them to buy their firms. And even if they had such credit, they might not want to own the firm simply because doing so is risky; it entails putting all one's eggs into one basket. In a more egalitarian economy, these problems could be smaller.
Now, these mechanisms - many of which are discussed more expertly in Sam Bowles' The New Economics of Inequality and Redistribution - will vary in strength from time to time and place to place. And no doubt, Ryan can think of potentially offsetting ones.
But there's one big fact which hints that they might be significant. Productivity growth has been much lower recently than it was in the 80s. This should be puzzling to people like Ryan, because for years they've told us that Thatcherite reforms in the 80s should have boosted productivity growth. So why has it fallen? Might it be that the benefits of those reforms have been offset by the fact that the increased proportion of income going to the 1% depressed productivity through the above mechanisms?
I can't prove this to Ryan's satisfaction. But shouldn't we at least take the possibility seriously?
Except that inequality is often compared against that in the US, the most unequal of the advanced economies and surely their productivity is substantially above ours, and has been for most of the time since these things were measured. Is the US just an outlier, a statistical anomaly or are there specific factors countervailing the above mechanisms and hence neutralising the negative effects of inequality?
Posted by: paulc156 | December 09, 2014 at 03:41 PM
the other side of 4. is that I imagine greater inequality is associated with larger, more concentrated 'bad' neighbourhoods, and you can tell stories about negative agglomeration effects, intergenerational transfer etc.
Posted by: Luis Enrique | December 09, 2014 at 03:43 PM
I have cited this paper here before, but one could extend the efficiency arguments about talented women and ethnic minorities being overlooked, from this paper
http://klenow.com/HHJK.pdf
to talented people born poor in highly unequal societies.
Posted by: Luis Enrique | December 09, 2014 at 03:55 PM
7. High inequality should correspond to a plentiful supply of low-paid workers, so investment in mechanisation/automation is lower.
Posted by: gastro george | December 09, 2014 at 03:56 PM
George makes a good point, plentiful supplies of cheap labour motivate capitalists to employ them over more productive machines.
But isn't the question itself wrong, who gives a shit about productivity when the products are going to a few? Less productivity and more equality could mean the majority get more.
Posted by: An Alien Visitor | December 09, 2014 at 05:35 PM
Might inequality depress entrepreneurialism among the non-rich and encourage rent-seeking among the rich?
Posted by: Martin S | December 09, 2014 at 11:51 PM
"....how does a larger gap of outturn income reduce productivity growth?". Indeed how? Whilst we do have income gaps and we do have reduced productivity I am not sure the one causes the other. Hard to see how a zero-hour delivery driver's productivity is much affected by her reading the FT and misery and job dissatisfaction is and was pretty common. The suggested mechanisms seem pretty weak to me.
The question arises 'what is so great about productivity (in the UK)' when seen from the viewpoint of a global capitalist. From this perspective the UK is a place to hire hot shot financiers and lawyers and a competent distribution network and nothing else. Given this it hires from the upper decile and the mid-range and ignores the rest. No need to train or improve either group - just hire better when needed. Essentially cherry-picking which suggests to me some element of old-style employment markets has disappeared. Given a government attitude - leave it to the market - the result is unsurprising - a market with big empty spaces.
Posted by: rogerh | December 10, 2014 at 08:09 AM
The latest evidence suggests inequality does not reduce trust: http://www.ncbi.nlm.nih.gov/pubmed/23347481
Posted by: Noah Carl | December 10, 2014 at 08:21 AM
@ Carl - Thanks for that. However, other international panel evidence suggests it does, at least for developed economies:
http://www.rcfea.org/RePEc/pdf/wp25_14.pdf
@paulc156 - this is a huge issue. Remember, though, that in the 50s and 60s (when the US's economic performance was good), income inequality in the US was not especially high.
Posted by: chris | December 10, 2014 at 12:13 PM
I think the US anamoly may be explained by the fact that the parts of the US which contribute to growth most are relatively more equal 'societies' than the rest of the country.
I don't have the stats to hand but the coasts contribute more to US GDP than the south. And certainly a lot more in terms of productivity.
I would be very interested to see if New York and California had more equal gini coeffs than Louisiana and Mississippi.
If we were to do a hypothetical experiment and lop off the coasts from the south we'd end up having one country looking a lot more like other Western countries and a Republican midwest and south which would have a lot more in common with Central Asia.
Posted by: Icarus Green | December 10, 2014 at 12:27 PM
More generally to the point of the article, I suspect societies that are highly unequal have an establishment that are a lot more able to stop the state from redistribution e.g. Thailand. This is a bad thing.
People tend to see redistribution as an efficiency and productivity killer. I have no idea why. Maybe at a micro level I suppose, but at a macro level its indisputable that societies that redistribute more tend to do better than those that don't (with the exception of communist countries where state ownership of everything kills off incentives completely).
Why? Here are some reasons I've considered:
1. Redistribution takes money away from those unwilling to spend or invest it and gives to to people with high marginal propensities to consume. The multiplier effects from this consumption stimulus boost economic activity. Much like a car, the engine is now running, and of course the car (economy) will eventually cover distance (economic development).
2. It eliminates poverty traps. People stuck in poverty cannot escape it even with hard work and the best of intentions. They need to be able to build up a stock of capital to leverage into a productivity enhancing investment like education or a tractor for their farm or something - this is obviously good for society as a whole. (Microfinance is an alternative to this).
3. Redistribution is a great way to raise money for more neutral public investments like schools, roads, bridges, R&D, university funding etc. You are simply not going to raise much money taxing the poor and the middle class. Thats why a lot of the tax take is collected from the rich - they just have more money. Rolling out free primary and secondary education to the peasants probably did more to boost long term growth than most other policies - that had to be paid for by the rich at the start.
4. Redistribution greatly reduces the risk of plutocracy. A political elite that caters to the interests of a minority obviously does not have incentives to improve outcomes for the majority, particularly when the interests of the minority clash with the majority e.g. climate change.
Of course, since the 70s we havent had any real redistribution. We've basically been taking money from upper middle class bankers and lawyers and giving it to the poor. The people that make most of their income from land, financial assets or rents hardly pay any taxes. Thats why offshore tax havens and places like Luxembourg exist where a 1/3 of the world's wealth lies. These could easily be closed down by politicians but the aristocracy would castrate them if they did. We need to tax wealth/property not income and eliminate the use of shell companies, offshore trusts and the like. The serve no social purpose other than to enable tax evasion. Its a joke.
Posted by: Icarus Green | December 10, 2014 at 12:55 PM
Not sure about the more equal US coasts theory. New York City and Silicon Valley must have most of the mega rich and the poorest get similar levels of welfare all over.
Posted by: davidjc | December 10, 2014 at 01:31 PM
Yesterday I was at a seminar about peasant technology and productivity in the High Middle Ages (1000-1300), which is a seriously unequal type of society. The speakers were arguing that peasant productivity was higher on their own smallholdings than on the lord’s demense that they were required to work on. The reason wasn’t just higher labour input on the smallholdings, but also more effective techniques. In particular, they were arguing that there were a lot of micro-developments of tools made “on the ground” by the people using them. Lords, who were just interested in grain crops that they could sell off for profit, didn’t much care about such hand tools. (They also didn’t think much of the peasants generally).
In an unequal society, the people at the top are more prone to thinking that they know everything and that the workers at the bottom are just stupid, ignorant and lazy. If most innovation and improvements in productivity in fact result from cumulative minor changes by those doing the work, you’re losing a lot of know-how by ignoring them. And that also fits in with the managerialism that says that you need to import overpaid consultants to tell you what your workforce could already say if you’d listen to them.
Posted by: magistra | December 10, 2014 at 02:17 PM
@magistra +1
c.f. also the classic Toyota method of bottom-up product development.
Posted by: gastro george | December 10, 2014 at 03:42 PM
1) calls to mind the Chicago Bulls basketball team in the 1980s and 1990s, i.e., the Michael Jordan years. The story is that it wasn't until Coach Phil Jackson was able to persuade MJ that he, MJ, had to get everyone on the team to engage and stop trying to do everything himself, that the team became dominant. Further, even after that, in games where MJ took over, it was clear that his teammates stood back, and did little, expecting him to do it more. Typically in those games, the team did not to so well.
(I lived in Chicago for most of the MJ years, so ... It was a good location in space-time to be a basketball fan.)
Posted by: marcel proust | December 10, 2014 at 05:02 PM
When a fairly sizable portion of your population is working 2 or 3 minimum wage part-time jobs, which are very difficult to juggle and leave people exposed to high risk of financial disaster, productivity is not going to be high. People's health is affected. People's mood is affected. Their ability to deal with the usual small misfortunes is greatly reduced. It's a scandal that in 2014 America that any adult working 40 hours a week can't afford a small apartment, a car, insurance and utilities.
Posted by: Mary | December 10, 2014 at 06:11 PM
George and the Alien are on the track here. Technology is the principal factor in productivity, compared to which trust, motivation, effort, etc are rounding errors. Introduction of more productive technologies is attractive when the cost of labor is relatively high, and when the fixed costs can be amortized over large production runs, i.e. there is a mass market, i.e. lots of well-paid workers. (Though I am using the language of goods manufacture, the same factors operate in the provision of services.) On the other hand, "who gives a shit about productivity when the products are going to a few?" - the rich don't buy yachts because they are cheap transportation (...positional goods, Veblen goods, take your pick).
(Later) Not being of the yachting set, I did a little looking around; found a yacht-show report in which all the builders were touting features. Only one mentioned production efficiency and automation; that one (Cheoy Lee) has a significant business in commercial vessels, where costs no doubt are a significant element.
Posted by: Ken Schulz | December 10, 2014 at 06:25 PM
On the technology point, there are some good papers coming out looking at how workers will resist adoption of more productive technologies when the distribution of the benefits of those technologies is opaque. For example, the work by Eric Verhoogen (and others) here on soccer ball manufacturing in Pakistan. http://www.columbia.edu/~ev2124/research/ACCKV20140528.pdf
Posted by: Simon Halliday | December 11, 2014 at 02:11 AM
@ Chris -- Interesting, thanks.
Posted by: Noah Carl | December 11, 2014 at 05:16 AM
Excellent post. My question is - if CEO pay must be high to get the best talent - and assuming talent is important up and down the organization - why is it not also important to pay more at the median even if this brings down top incomes?
For my money the puffery is coming from corporate flacks who beat their breast while saying "our people are our most important asset" and then F%^k them at every opportunity.
Posted by: Bob | December 11, 2014 at 11:28 AM