One of the architects of today’s big banks, Sandy Weill, who helped build Citibank into the behemoth it is today, has come out and called for breaking up the largest banks “so that the taxpayer will never be at risk, the depositors won’t be at risk.”

If only it were so simple.

Weill should remember the savings & loan crisis of the late 1980s and early 1990s. That crisis mostly involved small institutions, yet it cost taxpayers a lot and did significant harm to depositors. Perhaps Weill believes the 400+ small banks that have failed in the current crisis had little effect on the economy. While I admittedly haven’t run the numbers, it’s hard for me to believe that 400+ bank failures did not have some negative macroeconomic effects, in addition to being very expensive for the Federal Deposit Insurance Corporation.

In fact, I would argue that the single largest problem facing the banking industry before this crisis was a lack of geographic diversification. Our long history of extensive branch banking restrictions kept banks small and extremely vulnerable to local and regional downturns. Fortunately, we deregulated that area in 1994. The result has been a more stable financial system. Would Wells Fargo even be standing today if it had been limited to the California housing market (where Wells Fargo got its start)?

Weill needs to tell us, if we were to break up the banks, where exactly will that risk go? It isn’t going to just disappear. As I’ve argued elsewhere, one result of our small, fragmented 1920s banking system was the creation of Fannie Mae, the Federal Housing Administration, and the Federal Home Loan Banks. Need I remind Weill that the current bailout of Fannie and Freddie is at least $180 billion and counting, far exceeding the costs of all other rescues in the recent financial crisis combined? If we’d had bigger banks in the 1930s, we could have avoided the creation of such disasters as Fannie and Freddie and the FDIC (witness the stability of Canada’s diversified banking system, both in the 1930s and recently).

The most effective solution to risk-taking by big banks, as I’ve argued elsewhere, is to stop trying to micromanage what risk banks are taking and start pulling back their safety net. It is largely the moral hazard created by various government guarantees protecting “Too-Big-To-Fail” banks that caused the most recent crisis. It’s time to start reducing, if not eliminating, those guarantees. Ultimately, Too-Big-To-Fail is a political problem, not an economic one. The solution is to be found in limiting government, not the banks.