Every student who’s taken Economics 101 has learned that money has three main roles. It serves as a medium of exchange, a store of value, and as a unit of account. The three roles are interrelated; changes in the value of money in any of these roles matter for its other uses.

There is also more than one kind of money. A simple distinction can be drawn between cash on hand, and cash in the bank. The bank runs during the Panic of 1907, for example, were the result of rational depositors doing their best to convert cash in the bank to cash on hand, given their concerns about the value of the former.

Fundamental accounting textbooks appeared soon after the Panic of 1907, including Charles Ezra Sprague’s The Philosophy of Accounts and Henry Rand Hatfield’s Modern Accounting. Sprague and Hatfield each had significant banking experience, and each offered curious principles for the treatment of “cash.” Sprague and Hatfield preached that cash on hand and cash in the bank were both cash, by fiat, and should be combined in the same account, at the same value. Over time, generally accepted accounting principles enshrined this basic falsehood.

The National Monetary Commission recommendations also arrived on the heels of the Panic of 1907, and in 1913, an array of political forces gave birth to the new Federal Reserve System. Advocates were effectively saying that the new central bank solution would help enforce the equality of those two unequal things, cash on hand and cash in the bank.

Twenty years later, the nation suffered its worst banking crisis to date with the onset of the Great Depression, in part because the cure was part of the disease. Congress then grafted deposit insurance into the mix, again trying to inspire confidence that those two unequal things — deposits and cash on hand — were equal.

Despite — or perhaps because of — the government’s efforts to equate the two basic kinds of money, we keep getting our heads handed to us by our banking system. But in recent weeks, a healthy development has emerged relating to money’s role as a unit of account.

Meetings for the G-20 nations last month included significant discussion about banking system stability, including agreements relating to the seniority of large bank deposits. In turn, new rules slated to go into effect in the U.S. in 2015 are designed, at least in theory, to raise the costs of large bulk deposits for banks where they reside, and to instill more market discipline for uninsured deposits. The implication is that the value of bank deposit money may depreciate if the bank where the money resides experiences deteriorating credit quality.

As I wrote about 12 years ago, “accounting principles that pretend that risky things aren’t risky can couple with other policies to provide a formula for instability, subsidies, and bailouts.” Accounting standards that give carte blanche to “cash and cash equivalents,” assuming they are safe and hold unchanging value, are at odds with market discipline.

We are now picking up the pieces after the worst economic and financial crisis since at least the Great Depression. This crisis was sparked in large part by moral hazard associated with the government safety net for financial markets. The safety net helped reinforce dangerous assumptions that risky things aren’t risky.

To curb such assumptions in the future, the Financial Accounting Standards Board and the Governmental Accounting Standards Board should undertake projects explicitly incorporating risk into the valuation and reporting of “cash and cash equivalents.”

Bill Bergman is director of research for Truth in Accounting, a nonprofit based in Chicago. He previously served as an economist and policy analyst at the Federal Reserve Bank of Chicago.