In the last 15 years, Congress has made only one serious effort to address the problems with the government-sponsored enterprises (GSEs) Fannie Mae and Freddie Mac. That effort now appears to be on the verge of producing results.

The Housing and Economic Recovery Act of 2008 created a new federal regulator, the Federal Housing Finance Agency (FHFA), with expanded authority over the GSEs. The law gave the agency authority to place Fannie and Freddie into conservatorship (or receivership) and required it to establish capital requirements for the companies.

Fannie and Freddie have been in government conservatorship since 2008, but with an eye toward ending those conservatorships, the agency is now set to propose a new regulatory capital framework. It would make zero sense to release the companies without any capital, and conservatorship is not supposed to be permanent, so it’s about time.

New FHFA Director Mark Calabria wants to have the final rule published by the end of 2020, so the proposal will be out soon, possibly as early as this spring.

It is arguably the most important rulemaking the agency will undertake. Historically, the companies’ lack of equity capital fueled both their abnormal growth and the outsize risk they posed to the housing finance system, borrowers, and taxpayers.

From 1990 to 2004, for instance, Fannie Mae shareholders’ average annual return on equity was 24.5 percent, with relatively little variation from year to year. (Fannie’s financials over this period are available here, here, here, and here.) This is nearly double the typical return for banks, even though both banks and GSEs are in the business of funding home mortgages.

No, the GSEs did not gobble up the highest quality mortgages and leave only the poorest-performing/least profitable ones for the banks.

>>>Strict Bank-Like Capital Rules Needed for Fannie Mae and Freddie Mac

What explains the abnormally high returns, aside from the line of credit with the U.S. Treasury, is the radically lower (and weaker) capital requirements. With the full blessing of the federal government, the GSEs’ leverage has been at least 5 times higher than that of the largest U.S. banks.

In common terms, the GSEs were a profit machine—they had a federal license to borrow money to the hilt, thus “levering up” their shareholders’ profits from funding home mortgages. The problem, of course, is that this process increases the risk that a business will fail.

Congress addressed this issue in 1992, but it imposed low capital requirements relative to banks. Back then, interest rate risk was still a main concern, and the GSEs insisted that they were not exposed to the same sort of credit risk as banks. Essentially, the thinking was that the GSEs would not be materially impacted—if at all—in the event that borrowers had trouble paying their mortgages. That sort of trouble, supposedly, was only a problem for the old-school financiers known as banks.

Perhaps this was a reasonable way of thinking in 1992, but the last few decades have definitively exposed it as wrong. There is no longer any doubt that when people are unable to make mortgage payments it reverberates inside of the GSEs at least as much as it does inside of banks. Fannie and Freddie simply are not immune to credit risk.

So there is no objective reason to let Fannie and Freddie get away with a substantially lower equity requirement than banks. Calabria has wisely stated that Fannie and Freddie should “maintain capital levels commensurate with their risk profiles,” and “operate under essentially the same capital rules as other large financial institutions.”

This approach is long overdue. Combined, Fannie and Freddie have total assets of $5.5 trillion, roughly half the total assets of all eight U.S.-based global systemically important banks (G-SIBs). Unlike the G-SIBs, nearly all of Fannie’s and Freddie’s assets are concentrated in just one type of asset (home mortgages). Arguably, this concentration makes them even riskier than the banks. At the very least, then, Fannie and Freddie should meet the same capital requirements as the G-SIBs.

This new paper provides more details, but the largest U.S. banks have to meet an eight percent capital ratio, as well as additional capital buffers like the supplemental leverage ratio and the capital conservation buffer. These buffers effectively serve to increase the minimum requirement. On a risk-weighted basis, the G-SIBs have an average 14 percent ratio of equity to assets, so if regulators believe that figure is too high for Fannie and Freddie, they should publicly explain their reasons.

Separate from the required leverage ratio, the FHFA should require Fannie and Freddie to hold a portion of their guarantee fees as cash reserves. It is often glossed over, but equity requirements do not dictate that a company holds any type of asset, much less cash, to pay for future losses. The equity amount will absorb future accounting losses, but the company still needs cash to pay for those losses. No insurance company—even a mortgage insurance company—is allowed to operate without liquid reserves for this very reason.

If the FHFA wants to better protect borrowers and taxpayers from the kind of economic collapse that occurred in 2008, its next move is pretty clear. It should require Fannie and Freddie to operate with capital and reserves similar to those of the nation’s largest banks and mortgage insurance companies.