It is very simple. The Stiglitz argument is that, when information is imperfect, then government might improve the market, not that it necessarily will. To argue that government will always improve a market when information is imperfect assumes that governments are perfect, which is perhaps even less tenable a view than that markets are perfect.

Imperfect governments can make imperfect markets even worse. When that happens in the home loan market, I rant about it. Unfortunately, it is more the rule than the exception.

The most direct way for government to deal with the disparity between the information available to borrowers and the information possessed by lenders is for government to mandate that lenders disclose the information borrowers need. That should be a slam-dunk case for government intervention, but it isn’t. There are good disclosures and bad disclosures, and all too often the disclosures mandated by government are bad.

A good disclosure is one that is relevant to the borrower’s decisions, that is provided early enough in the process to be useful, that is consistent with and complementary to other required disclosures, and that is uncontaminated by garbage disclosures that are useless to the borrower but absorb his time and deflect his attention. Alas, there are all too few of these.

The federal government has been in the mortgage disclosure business for the past 30 years, with terrible results: garbage disclosures that are useless to borrowers, conflicting disclosures by multiple agencies involved in the process, disclosures so voluminous in total that borrowers cannot absorb them, and disclosures made too late to do borrowers any good. Maybe the new consumer protection agency created by Dodd/Frank will do it better — maybe.

Government intervention can also take the form of rules regarding what lenders and perhaps others cannot do, or must do, to resolve a particular problem that borrowers have. There are good rules and there are bad rules.

A good rule is one that accomplishes its objective at minimal cost. Bad rules don’t, for a variety of possible reasons: They don’t deal with the underlying cause of the problem, they are enforceable only by deploying an army of regulators, or they stimulate legal forms of evasion. One bad rule often leads to another bad rule.

A good illustration of good versus bad rules is dealing with the problem of excessive charges to borrowers for third-party services required by lenders, including title insurance, mortgage insurance, property appraisal, and closing services.

These services have always been overpriced because borrowers paid for them while lenders selected the service providers. Competition by third-party service providers took the form of services provided to the lenders who made the referrals, which raised the costs of the service providers, which costs were then passed on to borrowers in the price of the service.

A good rule for dealing with this problem is to require that all third-party services required by lenders be purchased by lenders, who would embed the cost in the price of the mortgage. Since lenders are knowledgeable purchasers and can purchase in bulk, the price of third-party services would drop like a rock.

A useful way to think about this is to reverse the scenario: Consider what would happen to the price of automobile tires paid by car buyers if, instead of being included in the price of the car, they had to be purchased separately from firms selected by the car dealer. Can there be any doubt that car buyers would pay more for the tires than they do now when the cost is included in the car price?

The proposed rule would accomplish the objective because it confronts the problem head-on by eliminating the referral power of lenders, and would require no costly enforcement system. Borrowers who were charged separately for third-party services would be unpaid enforcement agents.

But this rule has never been adopted. Instead, lawmakers fashioned a bad rule: They prohibited the payment of referral fees. Instead of eliminating the referral power of lenders, this rule attempts to police how that power is used, which is impossible without an army of enforcement agents.

Small players often ignore the rule against payment of referral fees, and large ones have fashioned legal mechanisms that get around it. Reinsurance affiliates that share the mortgage insurance premiums paid by the lender’s customers, and joint ventures with title companies who get their title business, are legal methods of paying referral fees.

Needless to say, prices paid by borrowers have not been reduced by the law prohibiting payment of referral fees that have not been legalized.

Why do we have such a bad rule when there is an obvious good rule available? Stay tuned.

Discounted mortgages written by Countrywide, once the nation’s largest subprime lender, were granted to a far wider group of Fannie employees than the four top executives executives whose preferential loans were previously disclosed, according to Countrywide documents provided to Congress under a subpoena.

Countrywide’s VIP section, established to handle preferential mortgages for favored customers, serviced a variety of Fannie employees who handled Fannie’s business of buying mortgages and selling mortgage-backed bonds. Recipients included an account manager, a lobbyist, underwriters, lawyers, a home loan manager, a sales executive and a credit risk manager.

The documents reveal that when Countrywide was depending on government-sponsored firms to finance billions of dollars worth of subprime loans that touched off the housing meltdown, it was giving employees at the largest of those companies — Fannie Mae — sweetheart deals on their own home loans.

Countrywide was acquired by Bank of America in mid-2008. The documents were turned over to the House Oversight and Government Reform Committee by Bank of America. The government seized control of Fannie Mae and its smaller government-sponsored competitor, Freddie Mac, in September 2008. So far, the takeover has cost taxpayers $145 billion and is likely to be the most expensive of all the financial bailouts.

Rep. Darrell Issa of California , the House committee’s senior Republican, said Countrywide’s preferential VIP mortgages for Fannie employees spiked in 1998, when Countrywide was negotiating volume discounts on the subprime mortgages it was selling, and again from 2001 to 2003, at the edge of a housing and mortgage boom.

In a letter to the Federal Housing Finance Agency — the government agency that regulates Fannie Mae and a smaller competitor, Freddie Mac — Issa said Countrywide’s 153 loans to 37 Fannie employees were part of a attempt to vastly expand business with Fannie to the detriment of Freddie. Though government-chartered institutions, both Fannie and Freddie were owned by private stockholders.

“In 1999, Countrywide reached an exclusive agreement to sell Fannie Mae billions of dollars in mortgages at a discounted rate,” Issa said in the letter.

Records compiled by a trade publication, Inside Mortgage Finance, show Fannie rapidly expanding its purchases of Countrywide mortgages and a decline in sales of them to Freddie.

In 1998, Countrywide sold $25.6 billion in loans to Fannie and $17.7 billion to Freddie. By 1999, the figures were $30.8 billion to $11.2 billion in Fannie’s favor. By 2004, the spread was much wider: $67.7 billion in Countrywide mortgages sold to Fannie Mae compared with $2.9 billion in mortgages sold to Freddie Mac.

Also among the subpoenaed documents was a May 2001 “confidential and proprietary” e-mail from a Countrywide official to other company officials discussing the sensitivity of the discounted VIP mortgage loan to Daniel Mudd, then Fannie’s vice chairman and chief operating officer. He later became chief executive.

“Make sure the branch and RVP understand the sensitivity of this deal,” the e-mail said. “We already are taking a loss, it would be horrible to add a service complaint on top and lose any benefit we generate.” The meaning of RVP is unclear.

Credit ratings agency Fitch Ratings projects that about two-thirds of borrowers with permanent modifications under the Obama plan will default again within a year after getting their loans modified.

Obama administration officials contend that borrowers are still getting help — even if they fail to qualify. The administration published statistics showing that nearly half of borrowers who fell out of the program as of April received an alternative loan modification from their lender. About 7 percent fell into foreclosure.

Another option is a short sale — one in which banks agree to let borrowers sell their homes for less than they owe on their mortgage.

A short sale results in a less severe hit to a borrower’s credit score, and is better for communities because homes are less likely to be vandalized or fall into disrepair. To encourage more of those sales, the Obama administration is giving $3,000 for moving expenses to homeowners who complete such a sale or agree to turn over the deed of the property to the lender.

Administration officials said their work on several fronts has helped stabilize the housing market. Besides the foreclosure-prevention plan, they cited government efforts to provide money for home loans, push down mortgage rates and provide a federal tax credit for buyers.

“There’s no question that today’s housing market is in significantly better shape than anyone predicted 18 months ago,” said Shaun Donovan, President Barack Obama’s housing secretary.

The mortgage modification plan was announced with great fanfare a month after Obama took office.

It is designed to lower borrowers’ monthly payments — reducing their mortgage rates to as low as 2 percent for five years and extending loan terms to as long as 40 years. Borrowers who complete the program are saving a median of $514 a month. Mortgage companies get taxpayer incentives to reduce borrowers’ monthly payments.

Consumer advocates had high hopes for Obama’s program when it began. But they have since grown disenchanted.

“The foreclosure-prevention program has had minimal impact,” said John Taylor, chief executive of the National Community Reinvestment Coalition, a consumer group. “It’s sad that they didn’t put the same amount of resources into helping families avoid foreclosure as they did helping banks.”