How much should workers get paid in a country going through (or aspiring to) broad-based economic growth? the economist asks. She quickly explains her use of should refers to an equilibrium solution rather than a moral sense of ideal, but her historian friend is still not convinced. What is relevant, the historian says, is why in some places and times in the past workers got a larger slice of the cake than in others. The economist scoffs at this: income distribution is not analogous to splitting one single cake. Economic growth can produce Pareto-efficient outcomes: if as a worker you eat more cake than before you should be happy, regardless of how much more cake than you capitalists get. And yet, the historian replies, people sometimes aren’t happy with that outcome—they care about relative slice sizes even if you can show them that there is more cake in the oven. The standards of material welfare of most workers improved in post-war England and Sweden, and quite impressively so. But Swedish workers were distributionally better off than England’s, as their wages grew faster than their employers’ profits. Why did workers in post-war Sweden get a larger share of the new cakes of the golden age of western capitalism than their English counterparts? And why, if at all, does this matter?
It wasn’t primarily about the general rules of the game in the labour market, as in both post-war England and Sweden widespread collective bargaining ensured high levels of non-market coordination in industrial relations. There wasn’t a policy gulf either, at least at first: Sweden became well-known internationally for its large and generous welfare state, but government spending in health, education, and pensions was also rapidly expanding in post-war Britain. Here, social security spending grew at a mean annual rate of 4.5 per cent between 1949 and 1996, outpacing average income growth by 50%. At 2.4 per annum, the growth in real disposable incomes was impressive itself: the mean Briton in the late 1990s was almost three times better off than fifty years before. So there was definitely more cake in the oven and there was an increase in public spending to try and guarantee that everyone got at least a slice. Yet inequality in Britain increased over the second half of the twentieth century, particularly over the 1980s when the number of people earning half the national median increasing from 5 million to 13.5 million.1
Despite some common goals in terms of policy shared by the likes of Clement Attlee and Tage Erlander, the historical starting point was very different with regards to the functional distribution of income. Bengtsson has shown that the Swedish labour share had surged already in the 1920s, well before collective bargaining and the onset of the Nordic welfare state, as a result mainly of low unemployment rates and high trade union density. Granted, and as Piketty argued for the advanced economies as a whole, the losses of the top income earners in twentieth-century Sweden were not the gains of the poorest Swedes, but of a ‘patrimonial middle class’, of which many unionized industrial workers became a part. The political power wielded by unions also affected the specifics of collective bargaining, particularly the level at which wages and industrial relations were negotiated. To put it briefly, in the UK it was always closer to the plant-level, or at best at the industry level, whereas the Swedish model relied on central wage-setting, often with effective sanctions for employers who failed to comply (see Golden et al’s comparative study).
Inequality between capital and labour (and within wage-earners as well) was, then, already much lower in Sweden than in Britain (and anywhere else for where we have data, according to Roine and Waldenström) by 1950. But during the expansion of the welfare state top incomes in Sweden continued to decline, which didn’t happen in Britain. This is explained by the comparative intensity of social spending.2 As Johnson has shown, in the era of expanding social spending in Western Europe (1960-1975) British public spending in education, health and pensions expanded at a slower rate than in the other European OECD economies, and since the 1980s it decreased as a share of GDP far faster. This gap across the English Channel can also be seen as a lag: by 1994 Britain was spending 5.4 per cent of its GDP in education, a figure similar to Sweden’s four decades earlier.
Post-war Britain’s workers were distributionally worse off than Sweden’s because labour’s share in Britain was much lower than in Sweden already before 1945, and a comparatively shorter and less intense welfare state expansion was not enough to revert that trend. This matters because labour’s share is a key driver of inequality—the incomes of labour (wages) are much more evenly distributed than the incomes of capital (profits)—and the British path of increasing welfare expenditure with increasing inequality is largely explained by rapidly declining labour’s shares of national income, especially since the mid-1970s, which were documented in detail by Ryan. Of course, capital played its game as well. The strong comeback of private capital in the richest countries since the 1970s which Piketty emphasised is the other side of this coin, and it affected Sweden as well as Britain. But if the trend was shared, its level wasn’t. As a result, today Britain and Sweden sit at opposite ends of the European continuum in terms of income inequality, with the top decile’s share of income at 40% and under 30% respectively. Perhaps that is one of the reasons why a Swedexit isn’t in the cards.
- Inequality in material welfare is of course not only about incomes, but about lifetime itself: by the late 1990s British male blue-collar workers lived 7.4 years less than their counterparts in “professional” trades (Department of Health, 2002).[return]
- And also, although to a lesser extent I think, by taxation, though we’ll leave that for the World Cup Finals (if, as it seems likely, England faces either France or Belgium).