The extension of the vote to almost all adult males by the late 19th century was followed by an enormous expansion of welfare provision by governments in Western Europe and North America. Government spending on healthcare, education, pensions, unemployment payments and other social transfers rose from zero in 1910 to about 35% of national income in the UK, France, Germany and Sweden by the mid-1970s, and has remained at that higher level. The US experience has been broadly similar. This is surely the biggest change in the nature of society since the nineteenth century: the one that our nineteenth century forebears, our great-great grandparents in my case, would find most important and most extraordinary.
To finance this seismic change, the share of national income going to governments in the form of tax revenues has increased over the same period, roughly the last 110 years, from around 10% to around 35% in the US, and up to over 50% in France and Sweden. (The UK and Germany lie in between at about 40% and 45% respectively.) Since governments spent about 10% of national income prior to 1910 on defence etc., governments ran mostly balanced budgets before the great expansion in welfare, and small but persistent budget deficits after. In consequence, government debt has increased steadily since the early twentieth century in the developed world.
There are two things about this great expansion in the role of the state which perhaps deserve more attention than they receive. The first is that most of the increase in welfare provision by Western democracies comes in kind rather than in cash, as pointed out by Emmanuel Saez at Berkeley University. There is variation in the breakdown across countries, obviously, with America more likely to give money to citizens to buy privately provided health services than Europe, but everyone provides public high school education directly, for example.
Second, neither the tax take nor the share of welfare provision, high though they are by historical standards, has budged much since 1980. (Pensions are the great exception, because the share of retired people in the population has risen dramatically since 1980s as people started living longer.) This is strongly suggestive that they can’t rise much further: that for whatever reason, tax-funded government welfare provision hit a binding ceiling in the late 1970s from which it can rise no further without further radical change in the nature of our society and economy, of the same order of ‘radical’ as the extension of the vote to all adults a century ago.
It's important to remember this fact when listening to political candidates of all sides. When British politicians promise or demand a huge increase in healthcare funding, American progressives demand a reduction in inequality, or fiscal conservatives insist that government debts must be reduced to more sustainable levels, it’s important to remember that any of these goals would require large and sustained increases in taxation. If the history of tax revenue’s share of national income since 1980 tells us much, I think it tells us that such further tax increases are politically impossible, or they would have happened somewhere by now.
Small countries can still increase their tax revenues at the expense of their bigger neighbours, of course. Ireland, Switzerland, Luxembourg and others can attract the mobile rich by offering them low personal or corporate tax rates, for now. But such solutions are not available to larger countries like the US, Germany, France or the UK.
Why is it that tax revenues appear to have topped out at between 35% and 50% of national income in the developed world in the past four decades? We can’t be sure of the answer, but one strong possibility is that otherwise successful people psychologically simply stop trying to succeed when faced with marginal income taxes much above 50 – 60%.
Income taxes are not the only taxes, to be sure, but in most developed economies they are the main source of government tax revenue, and in order to raise tax revenues in a meaningful way, most governments from America to Germany would have to raise income tax revenues. I’m convinced that raising income tax revenues much higher than they are now on a long-term basis just isn’t possible.
Why do I think so? Introspection, for one. Suppose you value an hour of leisure at about $500. If you find such a number implausible, ask yourself if you frequently or even occasionally spend that much on a restaurant meal or a bottle of fancy wine, or what is the maximum you would consider spending on a holiday for your family. Actually, $500 is pretty low for the kind of readership we want to attract. Now imagine you get a call asking you to give an after-dinner talk to a group of businessmen for an hour, with a fee of $1,000. (We can change the numbers. These ones are easy.) After tax, the fee offered only just beats your value of staying home and opening a bottle of wine.
So what happens in this situation? Either you ask for and get a higher fee which motivates you to accept the offer, or you turn it down. If the latter, then you didn’t get $500 after tax, but the government also missed out on $500 of tax revenue.
Critics of this argument point out that the speaking gig might be offered to someone else after you turned it down, who has a lower cost of leisure (perhaps they’re younger and poorer) and so the government gets its $500 after all. I think the critics might be right maybe 50% of the time, which still leaves the government out an average of $250. But the other times are times when nobody else but you will do, or nobody else who will do is available either. Then either speaking fees have to rise, or it's no speaks.
This argument combines with the important point that the income distribution is very skewed. Very roughly, the top 1% of income earners pay about one third of the whole income tax revenue. The next 10% pay another one third. That makes it not only hard, but quite risky, to try increasing income tax revenues much above their current levels. Indeed, current high marginal tax rates may even be set at levels that fail to maximize income tax revenues: lowering them would increase earnings rates by the highest taxpayers.
There is anecdotal evidence, for example from the UK, that lowering the highest income tax rates leads to an increase in tax receipts from the highest-rate tax taxpayers. I emphasize this is not, or not usually, because higher rate taxpayers are routinely evading tax, but because the alternatives to working hard and paying high rates of tax are either moving to another country with lower tax rates (something most people don’t really want to do – you really have to hate paying tax to want to live in the Isle of Man, for example) or better still, just not working so hard. The richest people tend to have the best alternatives to working: sailing, riding, writing, exploring the world, collecting wine or art, restoring a famous house or garden. That’s why it's so hard to tax them.
So if tax revenues can’t rise much in the rich democracies, what else could change that would allow politicians to deliver their programs? I’ll discuss my suggestions in future columns, but in the meantime please write in with your own ideas. But if I’m right, and this is so, this tax ceiling should do nothing less than the way we talk about both politics and economics.
Well the obvious answer would be the programs have to change -- But those undertakings also seem to be politically problematic (e.g. Social Security). I would say rationing becomes the end answer because as you have pointed out before, the political will of the people, at least in the U.S., is not to manage something until a crisis actually occurs and by then it is too late.