Wealth Accumulation over the Life Cycle and between Generations

En lång rad vetenskapliga studier har dokumenterat att personer med god ekonomi i genomsnitt är både lyckligare och lever längre än personer som har svårt att få pengarna att räcka till. Sambandet gäller även i nästa generation: Barn till förmögna personer klarar sig till exempel bättre i skolan och har bättre hälsa. Tidigare studier visar också att förmögna personer åtnjuter högre avkastning på investerat kapital. Detta innebär att rikedom kan fungera som en självförstärkande god spiral: rikedom kan leda till bättre hälsa, högre utbildning och högre finansiell avkastning, vilket i sig är viktiga förutsättningar för ekonomisk framgång. Men det är även tänkbart att fattigdom kan leda till motsvarande onda spiraler, svår skuldsättning kan till exempel göra det svårare för föräldrar att ge barnen en bra start i livet. I det här projektet planerar forskargruppen att studera sådana självförstärkande förmögenhetsbildningsprocesser empiriskt. En empirisk utmaning för att studera sådana processer är svårigheten att skilja på orsak och verkan eftersom de är intimt sammanflätade. Gruppen adresserar denna empiriska utmaning genom att använda variation i förmögenhet som är slumpmässig och därmed orelaterad till tidigare förmögenhet och utfall.

A tragicomical remark sometimes made is that it is “better to be rich and healthy than poor and sick”. This immediate truth hides a deeper insight: desirable outcomes are often positively correlated. One possible reason for such positive correlations is that good outcomes can be mutually self-reinforcing. For example, if you are healthy, you perform better at work and earn more, and because you earn more you may afford a healthier lifestyle. Another example is that you may put more energy into investing wisely if you are wealthy, which increases financial returns and boosts your wealth further. Similar mechanisms can be at play also across generations. If you are richer, you can afford to spend more time with your kids and invest more in their education, which in turn may improve their future income. 

If such virtuous circles, and the corresponding vicious circles, are important, inequality is self-generating, and a temporary shock to the wealth distribution may have persistent, and even growing, effects on inequality. It also becomes more difficult to address growing inequality in the presence of such self-reinforcing processes. On the other hand, if they are not present, temporary increases in inequality is less of a concern since they are unlikely to be persistent. This is one motivation for studying wealth accumulation over time empirically. 

Studying wealth accumulation empirically is challenging. The idea that cause and effect might be mutually reinforcing highlights the difficulty in distinguishing correlation from causation. In order to break the “causal chain” it is necessary to rely on some exogenous variation in wealth (or alternatively, the exogenous variation in e.g. educational outcomes). In this proposal we outline five different studies that exploit different sources of plausibly exogenous variations in wealth to study long-term wealth accumulation processes. The first four studies are based on three samples of Swedish lottery players who played the lottery between 1986 and 2011. The overarching question for these four studies is whether lottery winners invest the prize money in ways that increase their own wealth or the wealth of their children. In total, the lottery data encompass more than 400,000 winners and a total prize pool of more than six billion SEK. For each lottery, we effectively control for the number of tickets held by each player, thereby only using the variation in lottery wealth that is exogenous. In the fifth study we intend to use quasi-random variation in debt relief to study the potentially vicious circle by which indebted individuals may be trapped in poverty. Most of the data needed are already available to us, but we have not yet carried out any preliminary analyses. It might therefore turn out that some study will not be feasible, for example due to insufficient statistical power.