Why Mortgage Refinancing Is Not a Zero-Sum Game
In a recent speech, New York Fed President William Dudley called for actions “to see refinancing made broadly available on streamlined terms and with moderate fees to all prime conforming borrowers who are current on their payments.” This blog post explains how such a move could help stabilize the housing market and support economic growth. It also explains why mortgage refinancing is not—as some argue—a zero-sum game in which the benefits to one group are exactly offset by the costs to another.
Mortgage refinancing is one of the normal channels through which declining interest rates support economic activity, growth, and jobs. Falling mortgage rates reduce the amount of income households need to spend on servicing their home mortgages, freeing up cash flow for purchases of other goods and services.
For homeowners with adjustable-rate mortgages (ARMs), the required monthly mortgage payment declines automatically as the interest rate resets on the mortgage. This is a prominent and well-understood process in many other advanced economies—such as Great Britain—where ARMs make up the majority of the mortgage market.
For homeowners with fixed-rate mortgages—the vast majority of U.S. mortgage borrowers—the reduction in monthly payments takes place when the homeowner refinances the existing mortgage into a new mortgage at the lower prevailing mortgage interest rate.
When borrowers refinance and free up cash to spend, there will be an offset on overall economic activity as mortgage bonds are prepaid and investors in those bonds need to find alternative investments at precisely those times when other bonds are likely to offer a lower yield, reducing the investors’ income.
But we will argue that the offset is only partial. Why? There are two reasons. First, many mortgage bonds are held by government or foreign investors whose spending on U.S. goods and services does not depend to any significant degree on their income from the mortgage bonds. Moreover, the share of mortgage bonds held by such investors has increased. Second, the remaining, domestically based private investors are likely to cut back their spending much less than the borrowers raise theirs.
To better understand why the offset is only partial, let’s look at the figures in a bit more detail. As shown in the pie chart below, slightly less than 47 percent of agency mortgage-backed securities (MBS) are held by foreign investors and federal governmental institutions, including government-sponsored enterprises and the Federal Reserve. In these cases, we would not expect any domestic spending offset from a decline in the value of the MBS securities, for the reasons discussed above. An additional 8.3 percent of MBS are held by insurance and pension funds; for these funds, any spending effects are likely to be spread out over a relatively long period of time.
The fact that 55 percent of the investors are governmental, foreign, or long-term institutional holders suggests that each dollar of refinancing effectively translates into 44 cents of net increase in after-tax disposable income available for spending (= 0.55 x 80 percent, assuming a 20 percent marginal tax rate). If homeowners on average spend 90 percent of this additional disposable income, then each dollar of reduced mortgage payments would translate into 40 cents of additional spending.
However, this estimate of 40 cents per dollar is overly conservative, because the tendency for borrowers to spend out of increases in their disposable income likely exceeds the tendency for investor households to cut back spending in response to decreases in their interest income. For example, if investor households on average spend 70 cents of every dollar of after-tax investment income, then the overall impact of a dollar reduction in a borrower’s required monthly payment would increase by 7 cents [(=0.45 x 80 percent) x (90 percent - 70 percent)]—or from 40 to 47 cents. Taken together, these calculations imply that every dollar reduction in a borrower’s monthly mortgage payment stemming from a refinancing is likely to generate close to 50 cents of additional spending—a very different outcome than the absence of any change in spending implied if refinancing were a zero-sum game.
There is a second channel through which increased refinancing can support consumption: via its effect on house prices, wealth, and consumer confidence. Refinancing lessens the likelihood that a borrower defaults on a mortgage by creating additional cash flow that helps the borrower absorb any adverse income shock.
Thus, by reducing expected foreclosure sales, refinancing helps to restrain the downward pressure on house prices. Declining house prices lead to decreases in consumption by homeowners through a standard wealth effect, as well as the associated impact of house price expectations on consumer confidence.
Avoiding mortgage delinquencies can also avoid unnecessary loss of access to future credit, which could otherwise depress consumer spending for years to come and exacerbate challenges to fiscal rebalancing. For borrowers who go through a default and foreclosure, it often takes more than five years to rebuild their credit (Brevoort and Cooper 2010). During this period, the borrower’s consumption is reduced because of lack of access to credit, and when this behavior is widespread, aggregate economic activity declines.
Note that the effect of refinancing on consumption and jobs is shared broadly across the population of homeowners—including those that do not themselves refinance—and the economy as a whole. This improves economic outcomes for the nation and ultimately brings higher returns for investors as the economy expands and monetary policy is normalized. Thus, wealth and confidence effects also argue against the view of refinancing as a zero-sum game.
A number of commentators still argue that there is something unfair about refinancing. This is a strange objection to raise against mortgage refinancing specifically, since borrowers of all kinds—corporations and governments as well as households—refinance their debts when interest rates are low, freeing up cash flow to spend or invest. At the end of the day, investors and borrowers are parties to a set of contracts that accord this right of prepayment to the borrower, and the investor willingly accepts this risk—but at a price. Indeed, investors have been well compensated for this risk. Over the past two years, mortgage bonds have performed better than most models predicted because of greater-than-expected impediments to refinancing.
As we have shown, refinancing is not a zero-sum game and can benefit the entire economy. The greater the scale and scope of the refinancing program, the larger these benefits are likely to be. And these benefits are likely to be larger still if a broad-based refinancing effort is accompanied by other measures to help stabilize house prices.
*Joshua Wright is a senior trader/analyst in the Federal Reserve Bank of New York’s Markets Group.
The views expressed in this blog are those of the authors and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the authors.