Sometimes it seems as though no one debating housing finance reform was alive before 2008.

How the housing finance system operated then, and why it all went wrong, tells us a lot about the "reforms" that are in prospect.

For example, what can we learn from the fact that before the 2008 crash there was a "jumbo" market for loans bigger than the conforming limit? That was a fully private market, accounting for about 15% of all mortgages, which regularly priced mortgages at 25 to 40 basis points over the rate for Fannie Mae and Freddie Mac loans. There were studies by the Federal Reserve Board at the time that suggested such a spread was the result of the limited liquidity in the jumbo market, and if the jumbo market expanded the spread would narrow.

Yet now we hear from housing pundits that a fully private market—particularly the one that would be created by the House Republicans' bill —would result in a spread of 90 basis points over current rates. Huh? How would this happen, and why is the estimate so different from what the market actually produced when it was operating without government support?

Another example is the debate about the 10% private capital requirement in the Corker-Warner Senate proposal. Here, the history of the housing finance market has suddenly taken on an importance that it did not have when the issue was how a fully private market would behave. Some participants in the debate argue that the GSEs' losses did not exceed 4%, so the 10% capital requirement in Corker-Warner is too high. It should be more like 5%, if that.

However, what happened in the past on this question is of little value in determining what will happen in the future. The reason is simply that when the government controls a market—as it would under the Corker-Warner plan—it creates the future that it wants.

Even now, we are seeing the future that the Government Mortgage Complex wants. It includes a "duty to serve," which seems to be an expansion of the concept of disparate impact, but without the invidious element of alleged discrimination. Once the government is in control of the market, it will see to it that just about everyone who wants to buy a home will be able to get the credit to buy one. Unlike disparate impact, the concept has nothing to do with minority status; it appears to be simply an entitlement.

There are two corollaries to the duty to serve—"opening the credit box" and low-cost mortgages. Opening the credit box—a concept that is incomprehensible to anyone unfamiliar with housing market jargon—means reducing the underwriting standards to the point where borrowers who would normally be ineligible for mortgage credit will now be eligible. These borrowers are frequently described—recently in testimony by the Center for American Progress—as "all creditworthy borrowers."

If a borrower is truly creditworthy, the credit will be available, barring discrimination of some kind. If there is discrimination, it should be punished, but that does not tell us why a creditworthy borrower cannot get mortgage credit. The problem is clearly a difference of opinion between a lender and a borrower about who is creditworthy. The Qualified Mortgage devised by the Consumer Financial Protection Bureau defines a creditworthy borrower as anyone who can repay a mortgage at the time it is agreed, irrespective of the borrower's record of doing so in the past, or stake in the property (since neither credit scores nor down payments are considered necessary). The Government Mortgage Complex (GMC) is happy with that definition, and all the government's relevant financial regulators have signed off, even though they admit that mortgages meeting this test—had they been originated between 2005 and 2008—would have resulted in a mortgage default rate of 23%.

The other corollary of "duty to serve" is a low-cost mortgage, and government can certainly see to that. It would not be much of a reform, of course—nor would the duty to serve be achieved—if mortgage rates were so high that many would-be homeowners could not afford them.

The Corker-Warner proposal can accommodate this. Whether the private sector capital is 5% or 10% doesn't really matter. The amount of private sector risk will depend on the underlying mortgages. If the projected default rate is 23%, as it is likely to be under the prospective QM (and companion Qualified Residential Mortgage) rule, the return to the private sector will have to be extremely high. That will raise the costs of the underlying mortgages, probably far above the amount low-income borrowers could afford.

But government can and will fix that, too. The government will find a way to subsidize those mortgages, first by eliminating any payment to the proposed Federal Mortgage Insurance Corp., and later perhaps by "temporary" subsidies to the FMIC so it remains solvent, or an FMIC guarantee extended to the private sector participants, assuring them a "reasonable" return on their risk-taking that does not cause the cost of mortgages to rise excessively. A final expedient might be giving the FMIC the right to borrow with government backing—just "implicit," of course, because the debt is already too high.

In the end, we will once again have a system that looks suspiciously like Fannie Mae and Freddie Mac redux, with a housing bubble, a mortgage meltdown and a financial crisis. This is reform, GMC-style.

Peter J. Wallison is the Arthur F. Burns Fellow in Financial Policy Studies at the American Enterprise Institute.