How did inequality and household debt interact in the run up to the 2008/09 financial crisis? Today, a new report by NIESR for the Resolution Foundation provides new evidence on that question for the UK. The new analysis confirms the severity of the borrowing situation of low income households in Britain before the crash and raises difficult questions about patterns of consumption in an era of high inequality.
The report’s key contribution is to dig beneath headline figures for household debt to describe the borrowing picture for households at different points in the income distribution. It’s well established that UK household debt, in common with many other countries, ballooned in the late 1990s and 2000s, with the aggregate savings ratio—the percentage of household disposable income that is saved—turning negative in 2008 for the first time since records began. Yet so far these headline figures have been something of a black box.
Figure 1 from today’s report shows how the decline in the household saving position played out for households in different income deciles. It suggests that the poorest 10 percent of UK households saw their saving position deteriorate catastrophically from the late 1980s onwards, falling to a staggering negative 43 percent by 2007. Put another way, these households were outspending their incomes by 43 percent each year. Even for households on low to middle incomes (those in the second to fifth deciles) the picture was bad for much longer than was previously thought. On average these households had been outspending their incomes for anywhere between ten and 20 years by the time the 2008 crisis struck.
Source: NIESR analysis for the Resolution Foundation, FES
Some will dismiss these findings as unreliable and there is certainly good reason for caution on the specifics. But if the trends seen in Figure 1 do reflect the general pattern of UK spending and borrowing in the run up to the crisis, how should we interpret them?
Two main points of contention emerge between commentators, and although it’s far from definitive today’s report speaks to both. The first relates to the relative importance of income and consumption. Crudely speaking, two camps have emerged in the UK on this question. On the one hand, there are those who see the pre-crisis period as one of profligacy and spending sprees, with consumption soaring on the back of easy credit. On the other, there are those who tell a story of low income growth in which households in the bottom half were forced to borrow just to stay afloat.
Figure 2 speaks to this question. It shows how the two components of the UK savings ratio—consumption and disposable income—grew from 1997 to 2007. Certainly disposable income growth was shockingly weak for the bottom ten percent of households in this period, indeed official data suggests it was weaker still. We also know from wider work that income growth was weak or non-existent for low to middle income households in the later period from 2003 to 2008, supporting a weak incomes story.
Yet the figures on spending aren’t easy reading either. Consumption growth in the bottom half of households appears to have been surprisingly strong in this decade and even to have slightly outpaced consumption growth in the top half. We should be clear that this doesn’t mean low income households went on shopping sprees; we don’t know, for example, how much of this consumption was made up by the rising cost of essentials like food and fuel. And, importantly, we also don’t know how much of how much of this new spending in the bottom half went to service mortgages.
Source: NIESR analysis for the Resolution Foundation, FES
This gets us into the second big dispute between commentators: how much of a role should we assign the housing market in these trends? Big pre-crisis declines in the UK savings ratio would be much less worrying—or at least would be worrying in a very different way—if they were driven simply by increased mortgage repayments. After all, these can be seen as another form of saving, and a pretty sensible form in a booming housing market. In this case, Figure 1 would be little more than another aspect of the UK’s housing boom, and one we can be relatively sanguine about.
It’s hard to conclude either way on this front but the analysis does suggest that increased mortgage borrowing isn’t the only thing driving the figures for the bottom half of households. Across the bottom five deciles of UK household income, for example, the share of households with a mortgage isn’t particularly large, ranging from 10 to 24 percent. These proportions were also pretty stable in the decade before the crisis. At least in later years, then, this wasn’t a story of more low income people taking on mortgages (though there undoubtedly were big increases in the size of each mortgage).
Where does this all leave the link between debt and inequality? We should be careful about strong conclusions when so much relies on interpretation. Overall, though, it’s not hard to see something of a dynamic of ‘keeping up with the Joneses’—or, in technical terms, evidence for the relative income hypothesis—in the consumption figures above, a dynamic that would have realised itself in part in terms of home-buying.
In fact, if there was one moment of agreement at this morning’s launch of the new research it was over the risks that now face low income households in servicing the resultant mountain of secured debt. As Jonathan Portes, Director of NIESR, pointed out, in 2007 there were around 12,000 different mortgage products on the UK market of which around two thirds (nearly 8,000) were aimed at people with ‘impaired credit histories’. Today there are none of the latter, and though the debts they made possible appear serviceable for now, that could all change quickly when rates rise.
Has the rise in debt made households more vulnerable? OECD Working paper