Which is the greater number: $105 or $100 plus 3%? You might think this a trivial question, but about 60% of adults can't answer it correctly — and many of them still use financial services. It's an alarming situation. After all, smokers understand they're taking a risk, if only because they've seen the health warning on cigarette packs; but bank clients, especially in poorer countries, are often clueless about money matters.

Standard & Poor's Global Financial Literacy Survey goes further than any of the previous research in examining the phenomenon throughout the world. It's based on 150,000 interviews in more than 140 countries, all conducted last year. The study's financial literacy criteria were rather loose compared with the ones developed in 2008 by two of the leading lights in this academic field, Annamaria Lusardi, currently of Dartmouth University, and Olivia Mitchell of the Wharton School of Business.

Lusardi and Mitchell asked just three questions to test for a basic understanding of compound interest, the effect of inflation on savings and risk diversification. S&P kept these three areas, and added the simple interest question. To be deemed financially literate, a respondent needed to understand three of the four financial concepts.

The questions won't seem difficult even if you don't remember when you last had to do sums in your head. Yet not a single of the four concepts got a majority of correct responses globally. Even in advanced economies, fewer than two-thirds of respondents got each of them right. S&P's estimate of financial literacy throughout the world? A mere 33%.

That alarming finding is no surprise — previous studies have shown a low level of understanding of basic financial concepts, too. They have also found that those who are bad at answering the simple questions typically pay higher credit card fees or get fleeced by payday lenders and other financial vampires. The S&P survey's more important contribution is that it highlights the stark difference in literacy between the more and less economically developed countries:

[rebelmouse-image 27089671 alt=""finlit"" original_size="615x277" expand=1]

[rebelmouse-image 27089672 alt=""finlit2"" original_size="456x329" expand=1]

The understanding of basic finance is often about experience. That's why it's high among homeowners, who are often paying back mortgages and are forced to understand compound interest; and that's why in countries that have seen high inflation, such as Argentina or Bosnia, an inordinately high percentage of respondents got the inflation question right. People in wealthier countries, where the financial systems are well-developed and information is more readily available from various sources, including friends and neighbors, the experience is the most rounded. Accumulating it, however, is potentially costly, and it doesn't work for everyone. According to the S&P study, in the U.S., "3 in 10 adults with a housing loan are unable to perform basic interest calculations on their loan payments. Since the global financial crisis was triggered in part by mortgage defaults in the United States, this should concern policymakers, not just homeowners."

I'm inclined to agree that the gap between financial literacy and the penetration of financial services is a regulatory problem as much as one of education. No country allows someone who hasn't taken a driving test to drive a car on public roads. Yet all countries allow, even encourage their citizens, to hold banks accounts, take out loans and make investments, even though it has been shows countless times that many of them have no idea what they're doing and therefore could get hurt.

Introducing a basic literacy test, or questionnaire doesn't have to be a major hassle: The financial institutions could administer the tests by including them in standard applications. There should be more questions than Lusardi and Mitchell or S&P have been asking, because sometimes a product — say, a floating-interest loan or one issued in a foreign currency — requires more specific knowledge. If a prospective client fails the test, the law should require that the client demonstrate understanding before the application be approved.

Including tests in applications should be especially important to developing country governments. It would potentially save them a lot of trouble during financial crises: They wouldn't, for example, have to deal with tens of thousands of people holding loans in steeply appreciating Swiss francs, a problem Hungary and Poland recently encountered. People will often act irresponsibly when they want something badly enough, and it would be too intrusive for governments to try to rule that out. People may still make ignorant decisions, but it is the job of regulators to make sure they fully understand the risks.