Recent CEPR Discussion Paper No. 9238 (December 2012) titled "
Household Debt and Social Interactions" by Dimitris Georgarakos, Michael Haliassos and Giacomo Pasini looked at social determinants and drivers for debt accumulation amongst households.
According to the authors, "Debt-induced crises, including the subprime crisis, are usually attributed exclusively to supply-side factors. We examine the role of social influences on debt culture, emanating from perceived average income of peers. Utilizing unique information from a household survey, representative of the Dutch population, that circumvents the issue of defining the social circle, we consider collateralized, consumer, and informal loans. We find robust social effects on borrowing - especially among those who consider themselves poorer than their peers - and on indebtedness, suggesting a link to financial distress. We employ a number of approaches to rule out spurious associations and to handle correlated effects."
More specifically, the authors find that "the higher the perceived income of the social circle is, the greater is the tendency of respondents to take up loans and borrow sizeable amounts. This is true both for uncollateralized (consumer) loans and for collateralized loans…"
The above effect is "stronger for those who perceive themselves as having lower income than their social circle." In effect, this is keeping up with the Joneses effect, magnified by within-reference group peer effects.
"The tendency of households to take up uncollateralized and collateralized loans, controlling for the perceived average income of the social circle, is partly related to perceived spending ability or (computed) housing assets of members of the social circle."
"Moreover, we find that expectations about (the minimum) next period’s income are statistically significant for collateralized loans, pointing to a ‘Tunnel Effect’, but do not render perceived income of the peers insignificant. This is consistent with the idea that borrowing behavior is influenced by peer income not only because it conveys some information regarding the respondent’s own future, but also because of some comparison or envy effect." Notice - this is about basic human psychology, as co-determined by external (not internal or own-control) factors. In other words, any corrective policy will have to address the issue of peer effects, not only 'own effects'.
"Finally, the role of such comparisons is not confined to the tendency to borrow and to the level of borrowing conditional on participation, but it seems to extend also to financial distress."
To reinforce the argument above that the drivers of borrowing crises are social, not just individual (and hence any responsibility, liability and policy actions on this front have to be co-shared): "Our study has uncovered a potential for social influences on borrowing. By observing that others have higher average incomes, the household not only tries to emulate their
spending, as earlier studies have found, but also decides to borrow more, only partly because of expectations of higher future own income. Such decisions may be encouraged by a massive and unprecedented housing boom associated with high collateral values and expectations of continuing house price trends. The policy implication of our finding that social comparisons matter for debt behavior, after controlling for fundamental characteristics
of the household and region-time trends, is not to interfere with the process of forming social circles or perceptions regarding them, but rather to decouple perceptions of income or spending differences with peers from any decisions to borrow without proper account of the associated risks."
My view: let's cut puritanism bull**&t and recognise that debt crises are
not solely driven/caused by the reckless behaviour of individuals taken in an isolated setting, but are social / societal phenomena. This realisation should lead us to a recognition that dealing with prevention of future crises and with the fallouts from the current ones requires co-shared responsibility and liability.
Source: for earlier version (free to download)
http://arno.uvt.nl/show.cgi?fid=127996