In June, Bernie Madoff was condemned to 150 years in prison for what US federal agents have described as one of the biggest Ponzi schemes in history, which could leave investors with billions in losses. Swindlers like as Madoff are not new in history. Think about Ponzi himself, or the Victorian example of Merdle in Charles Dickens’ Little Dorrit. How could some of the world’s supposedly smartest investors of all time have been hoodwinked so easily? Our research could provide a possible answer.

Trust is considered a key ingredient in virtually all financial transactions. As stressed by Nobel Prize-winning economist Kenneth Arrow, “virtually every commercial transaction has within itself an element of trust… It can be plausibly argued that much of the economic backwardness in the world can be explained by the lack of mutual confidence".

Image: StockXpert

The intellectual tradition stressing the importance of trust goes back to at least Edward Banfield and Robert Putnam. In their influential books The Moral Basis of a Backward Society and Making Democracy Work, both authors show how civic attitudes and trust can account for differences in the economic and government performance between northern and southern Italy. Along these lines, sociologists, political scientists and, recently, economists, have argued—and shown—that having a higher level of trust can increase trade, promote financial development and even foster economic growth. Hence, the more trust, the better for a country’s economy.

But even if it benefits society at large, does it always pay an individual to trust? Old and recent financial scandals may raise doubts that this is actually the case. Those who trusted Bernie Madoff lost a lot of money and ultimately created little surplus.

To answer this question, we have examined the relationship between individual trust and individual income in our work. We argue that the relationship between trust and income is not always increasing—instead, it is hump-shaped. Perhaps surprisingly, not only those who trust too little but also those who trust too much do poorly, economically speaking. The individuals who do the best are those with more moderate opinions about other people’s trustworthiness.

Why? Our hypothesis is that people tend to think that others are like them when forming their beliefs about the trustworthiness of others. This results in two sources of suboptimal behaviour. On the one hand, individuals who are themselves untrustworthy, because they mistrust, tend to make decisions that are too conservative and, therefore, miss profitable opportunities. On the other hand, highly trustworthy individuals are too trusting and, because of this, get cheated abnormally often and incur large losses as a result. Somewhere in between these two extremes of trust, there is a “right amount" of trust that maximizes individual economic success.

The data support these intuitions. Individual-level data from the European Social Survey, administered in 26 European nations, reveal that individual income and individual trust are linked through a “hump-shaped" relationship. The cost of departing from the right amount of trust is quite large—it is of the same order of magnitude as the loss in earnings associated with forgoing college. Individuals who trust a lot tend to be cheated three times more often when dealing with a bank, and about two times more when buying goods second-hand, and when dealing with a plumber, builder, mechanic or repairman.

Yet, one could dispute all this evidence and argue that it may not be the reflection of wrong beliefs but rather trust capturing moderate traits that, as long argued by Aristotle, are key for successful performance. But this is not the case, since individuals who depart from the right amount of trust earn less even in an appropriately designed experiment where trust beliefs are isolated from the rest of the environment—and where the shortfall in earnings, in percentage terms, turns out to be similar to that observed in the survey data.

Does all this suggest that we should all be more careful in dealing with people in order to avoid being cheated, or that we should trust more to avoid missing opportunities?

Though both too much and too little trust are costly, the estimates show that, even for an individual, mistrust is more costly than excessive trust. Moreover, there is one important difference between the two extremes—while excessive mistrust and excessive trust are both individually costly, mistrust is also socially costly since it reduces investment and the creation of surplus. Compared with overly prudent parents, moms and dads who err upwards when instilling trust and teaching trustworthiness (the main mechanism through which these traits are acquired) not only improve the welfare of their offspring, but also raise that of society at large. For this, they should be subsidized.

Edited excerpts published with permission from Paola Giuliano is an assistant professor of economics at the UCLA Anderson School of Management and Luigi Guiso is a professor of economics, European University Institute, in Florence. Comment at