Mechanization, taxes and inequality

Google’s Eric Schmidt has told us that inequality is going to be the number one issue in the democracies in the coming years. Indeed, nearly every economic commentator tells us that rising inequality is not only the cause of our current problems but that as the robots come to take all our jobs this will only get worse. We must therefore tax the rich more in order to avoid a tragically fractured society.  

There’s only one real problem with this diagnosis: exactly what is being identified as the cause of the problem is what will solve said problem. 

The general analysis is that technological change, the way in which ever more jobs can be automated and done by software or machines, is going to lead to there being fewer jobs around for the workers to do. They will therefore languish on the scrap heap of unemployment, while the capitalists who own those new machines will reap ever greater portions of the income and wealth in the economy.

Cue increasing inequality until the proles revolt and slaughter their oppressors. The standard solution to this problem is to insist that returns on capital, and to those knowledge workers who also benefit, must be taxed more heavily in order to provide a dole for those displaced by the new machinery. That this analysis tends to come from those who already and always have supported higher taxation on the rich and greater income redistribution doesn’t actually make the analysis wrong.

However, there are reasons to insist that the analysis is indeed incorrect, whatever the motivations of those proffering that solution. The first and most obvious is that the economics of rising productivity simply do not work that way. Assume, as the argument does, that the rise of the machines leads to an increase in the returns on capital. OK: this is not just analogous, it is the same statement as the one that we have just created more capital.

In order to get any particular amount of whatever it is, we need to use less capital, for the productivity of capital has risen. We thus, for whatever capital stock we have, effectively have more of it. And we know what happens to the price of something that increases in supply: it falls.

Precisely and exactly because capital becomes more productive through mechanization, the price of it therefore falls. This is obviously the opposite of increasing wealth inequality across the society.

We can also check this assertion against history. The last time we radically mechanized a sector of the economy was agriculture between WWI and WWII. If it were true that the tractor and the mechanical harvester led to the capitalist making out like a bandit then all of the farmers would have become plutocrats. But they didn’t: they all went bust and have been supported out of public funds ever since. The reason being that by making the land more productive we effectively created more land, thus land fell in price, not rose.

So we would expect an increase in mechanization not to increase inequality but to reduce it. We get confirmation of this from Thomas Piketty’s new book on capital, which shortly after publication in English is being talked about as the great economic explanation of this 21st century. The usual suspects are, already, using it as justification for the usual solution, higher taxation and more redistribution. But those who are have failed to understand his basic argument, which is that it is economic growth itself that reduces inequality.

His argument is elegant in its simplicity. There is some rate at which already existing capital can earn profits and add to itself. There is also some rate at which wealth in general is being created in the economy. If that total growth rate is lower than the rate at which old capital grows then capital will become more concentrated, we have rising inequality. The reverse also applies. If wealth in general is being created faster than it accrues to the old fortunes then those old fortunes must be becoming a smaller part of the economy: inequality falls.

It is the implications of this argument that are so interesting. He points to the compression of inequality in the 20th century as proof of his argument. It was not union power, high taxation, or greater redistribution that led to inequality falling, rather, just that, especially post-WWII, the general economic growth rate was higher than the return to extant capital. Thus, voila, we find our societies becoming more equal. The slowdown in growth in recent decades is equally what has led to increasing inequality.

One fascination is that this obviates the need for the usual solutions proffered by those usual suspects. Since it wasn’t tax or the dole that reduced inequality last time around then it may well not be those that do it this time. What it was that did was swift economic growth: which is exactly the same thing as insisting that the robots are coming to take all our jobs.

For the mechanization of production is indeed a rise in productivity. This is also exactly the same thing as swift economic growth. So if it is true that the algorithms are about to computerize many work places, then this is the same statement as that we’re going to have swift economic growth which will reduce inequality. Exactly the thing that is being identified as leading to the problem, roboticization, is that very thing that solves that problem identified, inequality.

This does not mean that everything is going to be peachy of course, there’s still something that we as investors and business people must do, something that requires a certain amount of self-denial. As Charles Koch says in an excellent interview in the Wichita Business Journal:

“I think one of the biggest problems we have in the country is this rampant cronyism where all these large companies are into smash-and-grab, short-term profits, (saying) how do I get a regulation, we don’t want to export natural gas because of my raw materials … well, you say you believe in free markets, but by your actions you obviously don’t. You believe in cronyism. And that’s true even at the local level. I mean, how does somebody get started if you have to pay $100,000 or $300,000 to get a medallion to drive a taxi cab? You have to go to school for two years to be a hairdresser. You name it, in every industry we have this. The successful companies try to keep the new entrants down. Now that’s great for a company like ours. We make more money that way because we have less competition and less innovation. But for the country as a whole, it’s horrible.”

We all know how easy this is, a little lobbying here or there to make sure that regulation protects our profits stream and loads costs onto any would-be disruptive insurgent into our carefully managed and pruned markets. Plus the inordinate amount of time that we have to spend making sure that no one manages to use the politicians to get one over on us.

This is, to an economist, known as rent seeking and it’s also something that is injurious to swift economic growth. It is, as Koch says, just great for the company or producer that manages to carve out the opportunity, and bad for consumers and the economy as a whole. It’s true whether it’s that taxi medallion, a regulation that means no new plant can be built thus protecting the old ones grandfathered into the environmental standards (there will be no new copper smelters in the U.S. for example) or the rules in many States that you can only open up in certain industries if all of your potential competitors agree that you may.

These all restrict innovation, which itself delays or reduces productivity improvements and thus leads to slower economic growth, which, as we’ve been told, leads to increased inequality. That increased inequality leading to the siren calls that there must be that higher taxation upon us and more redistribution. All of which leads to a rather Marxian (i.e., informed by Karl Marx, not drinking the Kool-Aid) prescription for the future.

Marx himself predicted that us capitalists would band together, capture that state so as to make sure the rules were written to protect us and thus bring about that world of ever greater inequality leading to the revolution. And as Koch points out, there are indeed some who do capture the machinery of the state to do so, to collect those rents. But if we are to avoid that increasing inequality then we need to agree among ourselves to abjure the temptation to do so.


To retain that capitalist world that we so enjoy we need to ensure that it is also a free market one. For that is the only way to get the growth rate up and thus reduce the inequality that will, in the end, threaten the whole order.