Economists, especially the macro sort, look down on ordinary members of the public when they suggest that a nation’s finances operate in the same way as a household’s. They have a grand name for it: “the fallacy of composition”. And yet most of them suffer from very similar fallacious thinking. They imagine that running the production side of the economy is like running an individual business.
Economists are not good at looking beyond the aggregate statistics that describe an economy as a whole: total income (GDP or GNP), employment/unemployment, inflation, trade and current account deficits and so on. These numbers take on a reality in their own right, rather than being statistical abstractions. This leads to a ludicrously simple mental model of what is happening behind them. It’s a bit like understanding a car through the operation of the accelerator, brake, steering wheel and gear stick, without thinking about the engine. Economists imagine a national economy to be a bit like a single business, or perhaps an aggregation of similar ones, churning out all the things the people need. The more that gets produced, the more that gets consumed, and the happier we will all be, following the ideas of classical economics. Once you have reached full employment, the only way to increase production is to make the business more efficient by raising productivity. It is one of the central challenges to economic management in this classical view. And it is one Britain seems to be failing at. Commentators from right across the political spectrum (except the Greens, maybe) seize on poor productivity performance with their own favourite explanations.
But the economy is not analogous to a single business. To understand this we need to consider the different sorts of production activity that make up our economy. These divisions are largely my invention to convey the essential dynamics. Many actual activities are in two or three categories at once.
Let’s start with the easy bit: what I will call commodities. These go beyond globally traded things like oil and coffee to all goods and services that behave more or less as economists expect. Utilities might be an alternative name. These are most of the things sold at a supermarket, or basic cars, like a standard VW Golf, and the raw materials that make them. Services might include bank current accounts, household insurance and so on. Actual utilities a slightly more complicated, because they are distributed through networks that rarely compete with each other. All well and good, but what needs to be understood about these things is that from a consumer point of view there are diminishing returns. Your first fridge or car is really useful; your second one much less so. Not many people have three. That means demand is subject to saturation. Sales of white goods and cars rocketed in the 1950s and 1960s as people bought their first ones, but then slowed as most need came from replacements. This sea change for these and many other goods of the postwar light industrial revolution was surely one of the reasons for the economic wobbles in the 1970s, but one you will not hear mentioned by macroeconomists. A good definition of a developed economy is that it is one where most commodities have reached near saturation.
What are the consequences of this when we think about productivity? First, commodities are generally where advances in productivity have been the steadiest – production is relatively easy to automate and processes easy to redesign because there is little individuality in them. But as demand reaches saturation it means that workforces become smaller rather than more goods being produced. And that means that the weight of commodities in the total economy shrinks. First was agriculture, which used to account for the bulk of the economy, but now for about 1% or so. Next came manufacturing. And so on. How much of the total UK economy is in utilities? There’s no hard and fast definition, but it could be down to 20%.
Now let’s think about something else. Let’s call them “bads”. Direct bads are activities that are directly harmful; indirect bads are other activities we enter into to stop bad things from happening. Crime would be an example of the first, and police forces and security guards of the second. Into this category we might put armed forces and defence industries, a lot of the legal profession, regulators and compliance people, many consultants, and so on. We have little practical control over these activities, and they do not add directly to our wellbeing. An economy composed largely of security guards and armed criminals may have a decent GDP but it isn’t a good place to live. More is not better. The fewer bads overall, the better the economy will function for the population as a whole.
Productivity is clearly relevant for indirect bads. If you can treat the same danger with fewer warplanes or policemen then clearly that’s a good thing. But is this being measured in the economic statistics? And note that more productivity means the sector becomes smaller, like commodities but more so.
Next comes human services. The essence of these is human interaction. Economists’ favourite example is the hairdresser. The economy would be more productive if we all shaved our hair off, but somehow that misses the whole point. There are plenty of other examples: most of education, health care and social work, for example. One fascinating study (by BBC Horizon) showed that longer and more sympathetic consultations with doctors produced more effective treatments (the treatment in question was a placebo, which proved more effective than conventional drug interventions, but that’s an issue for another day). Productivity is a rather ambiguous concept for human services:if there is less waiting around and bureaucracy then that helps. But if you try to improve productivity by reducing the contact time with each client, you are likely to destroy the benefits of the service. This is not understood at all by economists. Trying to improve productivity in this way is how many public services are being undermined at the moment, with detrimental effects on society as a whole. But as a modern economy develops, human services, alongside hobbies and leisure, weigh higher, partly because we choose to consume more of them, and partly because there are few productivity savings to reduce the workforce. So the overall productivity of an economy (or at least the rate of productivity growth) will decline.
And then we have status goods and services. The primary purpose of these is to prove your status in society. The important point is for you own or use them when other people don’t. Think of luxury goods like smart sports cars or designer handbags. The interesting thing about these is that there is an anti-productivity factor. The more labour involved in producing the product, the higher status it confers. Improving productivity is a delicate matter. And, of course, status is a zero-sum game. Rising volumes of status goods simply lead to an arms race of expensive products that does nothing to advance human wellbeing overall.
I could go on. There are rents, public goods and finance and so on, each of which produce a different twist to the productivity puzzle, but none of which follow the classical pattern of commodities. There are two key points to make. First, improvements in productivity in a modern economy do not lead to expanded production, but to a redeployment of the workforce. That redeployment usually goes to sectors with lower productivity (status goods or human services), which means that the benefits of the original productivity gain are limited. But human wellbeing could well advance faster than overall growth, if people have more time for life-enhancing human services, for example. This isn’t a problem, it is a sign of a mature, successful economy. The second issue is that most economic activity is now in areas where productivity is practically unmeasurable because the outputs are intangible (bads, human services and status goods for example).
Look a bit deeper into Britain’s productivity statistics and both of these become important. The poor productivity growth is now attributed to a “tail” of small businesses, just as you would expect if productivity savings are being deployed into status goods and human services. Meanwhile the two sectors where virtually all measured productivity growth has happened in the last two decades are finance and business services. These are both sectors plagued by bads, the undermining of human services and rents. It is hard to argue that this growth has advanced human wellbeing. All of which leads to an alternative explanation of Britain’s low productivity. The British economy is simply further along the development path than others. In particular, unlike the Germany and France, we have run down export-focused commodity production, and that distorts any comparison.
There’s a further insight from this way of looking at things. Advancing human wellbeing in a developed economy does not come from producing ever more commodities. It derives from producing fewer bads, containing status goods, and expanding human services. Depending on how pricing works, that is unlikely to lead to measured growth in GDP. But that is not a bad thing: if it simply arises from the freely made choices of empowered citizens.
Of course productivity is an important issue for individual businesses and public agencies – at the level where managers should know whether value is being created or destroyed. But macro-economists should take their own advice on the fallacy of composition: don’t try running a national economy as if it was a business.