Building a Better Bailout
It's time to reward virtue.
Dec 1, 2008, Vol. 14, No. 11 • By LAWRENCE B. LINDSEY
Under my 4 percent mortgage plan, any bank that had to shrink its balance sheet would have a very easy way of doing so as loads of loan repayments would make the bank cash rich. It would no longer have to sell good assets and shrink its loan portfolio. If the amount of repayment exceeded the amount that the bank had marked down its mortgage portfolio, moreover, bank capital would expand rather than contract. The bank could then expand its portfolio and make even more loans if it chose to do so. TARP was supposed to do just this, but it didn't as the only information on pricing securities came from those same computer-generated models that misestimated the size of the mortgage problem from the start. Under this plan, virtuous homeowners who actually know they will stay in their home determine the price, and determine it with certainty, rather than relying on some computer model that, frankly, has no idea.
As noted in an earlier article in THE WEEKLY STANDARD ("High Anxiety," September 29, 2008), the underlying problem in the mortgage market is that the credit terms for home mortgages shifted abruptly from the most generous in history to more-restrictive-than-normal between mid-2006 and late 2007. The result was a drop of more than half in the demand for new mortgages (including refinancing) as measured by the dollar volume of mortgages at a time when there was an excess supply of roughly 3 million homes. In such a situation, prices drop.
The effects of the 4 percent mortgage plan on the housing market would be indirect, but quite real. The first effect would be to leave more people in their homes than would otherwise be the case. With an estimated 18 percent of all mortgage holders now in homes with mortgages that are higher than current market values, providing incentives for people to stay in their current homes is the best way of stopping still more excess supply from coming on the market.
But the government could easily magnify this effect by adding one more change in the mortgage terms: Allow the new mortgage on the house to be assumable. Under this plan, a buyer of a home with a new 4 percent mortgage gets to take over that mortgage as a part of the purchase. This substantially lowers the cost of acquisition and makes the house a far more liquid asset than it other-wise would be. Making a new 4 percent mortgage on an 80 percent loan-to-value mortgage assumable is equivalent to lowering the lifetime carrying cost of buying a home by 20 percent--compared to a mortgage rate of 6 percent. Alternatively, in a market in which all houses now have newly created assumable mortgages, the equilibrium price of homes would rise by 26 percent.
On the positive side, this onetime refinancing will not create a new bubble. New mortgages will not get the new terms, only existing mortgages. So while the plan will stabilize and possibly even increase the price of existing homes with mortgages, the effect is finite. It would take a sustained reduction in rates that applied to new mortgages to produce the financial fuel for another housing bubble.
The flip side is that a refinancing of existing mortgages is unlikely to revive the home construction industry. Homes yet to be built do not have mortgages, and so there is no existing mortgage to roll over under the new terms. This is unlikely to make the plan popular at the National Association of Homebuilders, but we still have a hangover of at least 3 million empty homes. Encouraging the building of new ones at this point would only delay the recovery of the housing market and the relief that is needed for the financial system. Stabilizing existing home prices and providing for financial recovery are, of course, the preconditions for a return to a vibrant home construction industry. So the help for homebuilders in this plan is still there; it is just a matter of timing and prioritization.
The American economy is in the midst of the fastest decline in consumer spending since 1980. The reasons are clear. Households are greatly overextended, having taken advantage of many years of very easy consumer credit conditions. The typical American household, for example, has 1.9 vehicles and 1.75 drivers. Credit is now being cut back drastically. In addition, rising unemployment is putting a crimp on incomes and creating caution among those with jobs.
Much of this is the inevitable reaction to excess. But it is also widely accepted that government has a role in making sure the adjustment process does not happen too fast or the cuts become too deep. Hence we hear proposals for more stimulus packages to put money into the pockets of consumers. During the campaign, President-elect Obama called for giving an average of $700 to middle-income families, making up the cost by raising taxes on upper income households.
A refinancing of home mortgages along the lines I am describing would be a much more dramatic stimulus. First, a family with a $200,000 mortgage at 6 percent (typically a family with an income in the $40,000 to $50,000 range) would receive an improvement in their annual cash flow of $3,000, four times as much as the proposed Obama stimulus payment. It is true that the Obama tax cut would also go to people without mortgages. But it is those with mortgages that are the most impacted by current credit conditions. Indeed, a refinancing on this magnitude is far better targeted at those most likely to respond to improved cash flow than are the proposed tax rebates.
The refinancing would also mean a permanent improvement in household financial conditions while a typical onetime stimulus package would not. The lesson from the 2008 stimulus package, and indeed from all other temporary tax cuts, is that the great majority does not enter the spending stream. Refinancing is likely to provide a permanent economic boost.
The refinancing is designed to be roughly budget neutral for the government. Currently the government can borrow for 10 years at about 3.25 percent and under 4 percent for 30 years. Mortgages are typically priced off the 10-year bond. In essence, the government would be borrowing and lending at the same rate. Those especially concerned with budgetary cash flow might prefer a fixed rate loan of 4.25 percent or might also consider putting "points" on the mortgage. As long as the rate remained substantially below current mortgage rates and homeowners were able to roll any points into the principal of their new mortgage, the impact on the incentives to take up the new program would be minimal.
Government would receive one further benefit. As mortgage payments dropped, so would the revenue loss from the mortgage-interest deduction. I estimate the extra revenue from this feature at between $15 billion and $20 billion. As lower income homeowners tend not to itemize and higher income homeowners face an increased tax rate, the distributional consequences of this feature would mean that most of the extra revenue would be collected from higher income homeowners. On the other hand, the prepayment wave would reduce the interest income on the mortgage portfolio held by the GSEs.
The big lesson of the bubbles of the last 20 years is that there is no free lunch. We are now paying the price of both the dot-com bubble and the housing bubble. But there are no free bailouts, either. A wave of refinancing on this magnitude carries a price tag. Done the way it is described here, roughly $9 trillion of mortgages would be refinanced. That is roughly 15 percent of total personal wealth in the country, clearly a huge undertaking.
But it is also worth bearing in mind that this figure exaggerates the true scale of what is happening. This is not a $9 trillion increase in the nation's indebtedness. It is the swapping of $9 trillion of one type of mortgage for $9 trillion of another type of mortgage. There is no net increase in the nation's debt.
There is an increase in government debt. But this is offset by an equal increase in government assets. To begin with, roughly $5 trillion of the mortgages to be refinanced are already on the government's books because of Fannie Mae and Freddie Mac. Funds have to be raised to issue the new mortgages, but each time a new mortgage is issued an existing mortgage of an equal amount is redeemed. The holder of the mortgage, i.e., the lender, bought that mortgage in search of long-term dollar denominated debt. When they are repaid, they will have to do something with the money. The vast bulk of the funds are also likely to be reinvested in long-term dollar denominated debt--particularly long-dated U.S. treasuries. Therefore, the net pressure on the financial markets will be fairly small even though the volume of transactions will be quite large.
There is a reason that government efforts have so far been focused on helping out those who have failed rather than those who have behaved virtuously. The left professes a belief in helping the needy. The right seeks to minimize government involvement and therefore compromises, agreeing to help only the needy. During the present crisis, the term "needy" has taken on new dimensions as some of the largest financial institutions in the country are at risk.
It is also far from clear that those who were over-extended on their homes, needy by one definition, are also worthy of aid. With regard to real estate, hard-working families can behave prudently while trust fund babies can behave frivolously, and vice versa. The problem is sorting out the prudent from the frivolous. Targeting relief on the frivolous induces people on the margin to behave frivolously. By contrast, targeting relief on those willing to assume full responsibility for their debts in return for a lower interest rate induces people to behave virtuously. A shift toward rewarding virtue would be the quickest way out of the debt morass we now find ourselves in.
Lawrence B. Lindsey, a former governor of the Federal Reserve, was special assistant to President Bush for economic policy and director of the National Economic Council at the White House. His most recent book is What a President Should Know . . . but Most Learn Too Late (Rowman and Littlefield).