Good morning. I'm William Dudley, president of the Federal Reserve Bank of New York. I would like to welcome all of you to the Bank and thank you for participating in today's workshop. My colleague Eric Rosengren, president of the Federal Reserve Bank of Boston and I asked our staffs to organize this workshop because the topic of the primary-secondary mortgage rate spread is important for current efforts by the Federal Open Market Committee (FOMC) to foster faster economic growth.
As I have said on a number of occasions, while the U.S. economy has expanded since mid-2009, the pace of the economic recovery has been too slow to make rapid inroads into the spare capacity left over from the recession. As a result the unemployment rate remains unacceptably high. To support a stronger economic recovery in the context of price stability, the FOMC took action at its September meeting to provide additional monetary accommodation. This included a new program to purchase an additional $40 billion a month of agency mortgage-backed securities (MBS). The FOMC reiterated in October that it will continue its purchases of agency MBS, undertake additional asset purchases, and employ its other tools as appropriate until there is a substantial improvement in the outlook for the labor market in a context of price stability.
Monetary policy works in large part through its effect on financial markets and financial conditions. The impact on economic activity depends on the extent to which policy actions are effectively transmitted to key sectors of the economy. While actions such as MBS purchases help ease financial conditions more broadly, a key channel through which monetary policy affects the economy is through the housing and mortgage markets. An easing of policy lowers the secondary market rate at which mortgage-backed securities trade, which puts downward pressure on the primary mortgage interest rate. A lower primary mortgage rate enables borrowers to refinance their existing loans at lower rates, and increases demand for home purchases, supporting house prices and household wealth, which in turn may increase household access to credit.
We have seen this channel in operation since the September FOMC meeting. Yields on newly issued agency MBS have declined about 45 basis points and the Freddie Mac survey 30-year primary rate has declined 23 basis points to 3.32 percent. These declines have been even larger if measured from the release of the minutes of the previous FOMC meeting in August, after which market expectations of easing in September increased. This is solid evidence that our policy has been and continues to be effective—though it is certainly not all-powerful in current circumstances.
However, the impact of monetary easing on the economy through housing and mortgage finance has been impeded to some degree by two factors. First, not enough people can take advantage of today’s low primary mortgage rates. Mortgage underwriting standards have tightened quite dramatically and the share of new mortgages going to households with better credit metrics has risen. While it is difficult to disentangle how much of this is due to supply versus demand factors, it appears likely that credit standards for home purchase loans have not just tightened since the housing bust but overcorrected.
In addition, many homeowners find it difficult to refinance their existing mortgage because they have little equity in their homes, or are outright “underwater” with mortgage balances that exceed the current appraised value of their property. In a speech earlier this year, I called for measures to ease refinancing and tackle other frictions and market failures that lead to economically inefficient outcomes in housing. Some useful steps have been taken. But, more could be done, and these issues warrant ongoing focus by policymakers.
Today, though, we are focusing on a second impediment to the impact of monetary policy on the economy through housing and mortgage finance: the significant widening of the spread between yields on mortgage-backed securities in the secondary market and primary mortgage rates. Actions taken by the FOMC such as its MBS purchase program operate principally on the secondary rate. For these actions to achieve their full impact, reductions in the secondary rate need to also pass through to the primary rate. To the extent that the primary-secondary rate spread widens the reduction in pass-through limits the full impact of the policy actions.
In the late 1990s and early 2000s the primary-secondary spread was in the range of 30 to 50 basis points. In recent years and over the past 18 months in particular, the spread has increased substantially. Following the September FOMC meeting it rose above 150 basis points. A spike in the spread following a decision to ease monetary policy is not surprising because the primary market may be slower to react. The spread has since retraced to its pre-September FOMC meeting level of about 120 basis points. However, the spread remains elevated by historic standards.
An important objective of this workshop is to gain a deeper and clearer understanding of the determinants of the primary-secondary spread and its dynamics over time. This spread is influenced by a number of elements, such as the valuation attached to the right to service the mortgages underlying the security, and the annual guarantee fee that is paid to cover the credit guarantee provided by the government-sponsored enterprises (GSEs).
The observed widening of the primary-secondary spread likely reflects many factors. Part of this widening is due to higher annual agency guarantee fees which are a necessary and overdue re-pricing of the credit guarantee provided to investors. While these guarantee fees differ by mortgage and by seller, the average effective guarantee fee has increased from 20 to 25 basis points to around 50 basis points today. But this still leaves a significant part of the spread left to be explained by other factors such as changes in originators’ costs and profits.
As you all know, the primary-secondary spread is a technically imperfect measure of the pass-through from primary rates to secondary rates. A new study1, by authors including New York Fed staff, adjusts for these factors to come up with a measure of originator2 profits and unmeasured costs, or OPUCs. The basic picture remains the same, with OPUCs at historically elevated levels.
The paper evaluates a number of potential factors that could help explain this increase. These include the decline in the value of mortgage servicing rights, credit losses and other costs associated with put-back risk, possible changes in pipeline hedging costs and other loan production expenses. It finds a potentially material role for the decline in the value of mortgage servicing rights in particular, but concludes that the increase in the aggregate measure of profits and unmeasured costs is “not likely to be driven exclusively or even mostly by increase in costs.”
This suggests that originator profits may have increased. The study examines a number of potential explanations. These include capacity constraints, market concentration, pricing power over Home Affordable Refinance Program (HARP) refinance loans and pricing power on other loans. It finds that capacity constraints play a role and that there is evidence to suggest that originators enjoy pricing power and elevated profits on HARP and other refinancings.
I encourage everyone to take a look at this paper. It represents a good starting point for today’s discussion. However, as the authors are the first to acknowledge, the increase in the primary-secondary spread and the underlying changes in originator costs and profits remain something of a puzzle. I hope today’s discussion will further deepen our collective understanding of this important issue.
To sum up, for monetary policy to be as effective as possible in supporting economic recovery in a context of price stability it is imperative that the key channels of the monetary policy transmission mechanism are operating as effectively as possible. The financial crisis and the housing bust created headwinds to the recovery in part through adverse impacts on the mechanisms of monetary policy transmission. The topic of today’s workshop is important for understanding how to gauge and potentially improve the transmission of monetary policy to the household sector. It is important that we correctly identify and to the greatest extent possible quantify the forces that have been acting on the primary-secondary spread. This requires bringing a broad range of informed perspectives to the table, perspectives that many of you have. For this workshop to succeed, we need for you to be as open, frank and precise in your discussions today as possible. President Rosengren and I thank you for your participation in this effort and we look forward to the key findings from the workshop.