«Elena Loutskina University of Virginia, Darden School of Business & Philip E. Strahan Boston College, Wharton Financial Institutions Center & NBER ...»
Other states followed suit, and state deregulation of intra-state banking was nearly complete by 1992 (Jayaratne and Strahan, 1996; Jayaratne and Strahan, 1998).
The transition to full interstate banking and branching was fostered by passage of the Interstate Banking and Branching Efficiency Act of 1994 (IBBEA), which effectively permitted bank holding companies to enter other states without permission and allowed banks to operate branches across state lines (Rice and Strahan, 2010). With these legal changes, banks now
across geographical markets through banks’ internal capital markets (Houston, James and Marcus, 1997).
Despite the passage of IBBEA, states continue to exercise authority under this law to restrict or limit interstate branch entry. While IBBEA opened the door to nationwide branching, it allowed states to influence the manner in which it was implemented. States that opposed entry by out-of-state banks could use provisions of IBBEA to erect barriers to some forms of out-ofstate entry, to raise the cost of entry, and to distort the means of entry. From the time of enactment in 1994 until the branching default "trigger date" of June 1, 1997, IBBEA allowed states to employ various means to erect these barriers. States could set regulations on interstate branching with regard to four important provisions: (1) the minimum age of the target institution, (2) whether or not to permit de novo interstate branching, (3) whether or not to permit acquisition of individual branches rather than whole banks, and (4) how tightly to control the percentage of deposits in insured depository institutions controlled by any single bank or bank holding company. Following Rice and Strahan (2010), we use these four state powers to build a simple index of interstate branching restrictions across states. The index equals zero for states that are most open to out-of-state entry. We add one to the index when a state adds any of the four barriers just described. Specifically, we add one to the index: if a state imposes a minimum age on target institutions of interstate acquirers of 3 or more years; if a state does not permit de novo interstate branching; if a state does not permit the acquisition of individual branches by an outof-state bank; and if a state imposes a deposit cap less that 30%. So, the index ranges from zero to four. We use this index below as a policy instrument to help explain variation in our key measure of financial integration.
The move toward integration in mortgage lending occurred in concert with branching deregulation, initially spurred by the activities of the Government-Sponsored Enterprises (GSEs)
- The Federal National Mortgage Association (Fannie Mae) and the Federal Home Loan Mortgage Corporation (Freddie Mac). By the 1990s, both Fannie Mae and Freddie Mac had become heavy buyers of mortgages from all types of lenders, with the aim of holding some of those loans and securitizing the rest. Together they have played the dominant role in fostering the development of the mortgage secondary market. As shown by Frame and White (2005), the GSEs combined market share has grown rapidly since the early 1980s. In 1990 about 25% of the $2.9 trillion in outstanding mortgages were either purchased and held or purchased and securitized by the two major GSEs. By 2003, this market share had increased to 47%.3 This market share fell after 2004 in the wake of the accounting scandals and the growth of subprime mortgages by private lenders, and then increased significantly since 2006 in response to the credit crisis. GSE access to implicit government support allows them to borrow at rates below those available to private banks, and to offer credit guarantees on better terms than competitors without such implicit support.4 As shown in Loutskina and Strahan (2010), the GSEs enhance mortgage liquidity, reduce the cost of borrowing, and increase mortgage acceptance rates conditional on borrower credit
3 GNMA provides a very important source of mortgage finance to low-income borrowers, holding or securitizing about 10% of all mortgages outstanding.
4 Passmore, Sherlund and Burgess (2005) argue that most (but not all) of the benefits of GSE subsidies accrue to their shareholders rather than mortgage borrowers. To take advantage their low borrowing costs, during the 1990s the GSEs increasingly opted to hold, rather than securitize, many of the mortgages that they buy. Policymakers became concerned about the resulting expansion of interest rate risk at the GSEs (Greenspan, 2004), although the 2008 crisis resulted more from the credit guarantees offered by the agencies than from exposure to their retained mortgage portfolio.
GSEs buy mortgages, they bear both credit and interest rate risk. When GSEs securitize mortgages, they either buy them and issue mortgage-backed securities (MBS), or they just sell credit protection to the original lender. In the first case, the originating bank retains no stake in the mortgage. In the second case, the bank continues to fund the mortgage and bear the interest rate risk, but obtains the option to sell the mortgage off as an MBS (because of the credit protection). In all cases, the GSEs enhance liquidity and thus foster integration of credit markets.
The GSEs operate under a special charter, however, that limits the size of mortgages that they may purchase or securitize. These limitations were designed to ensure that the GSEs meet the legislative goal of promoting access to mortgage credit for low and moderate-income households. The GSEs may only purchase non-jumbo mortgages, defined in 2006 as those below $417,000 for loans secured by single-family homes. The loan limit increases each year by the percentage change in the national average of single-family housing prices during the prior year, based on a survey of major lenders by the Federal Housing Finance Board. The limit is 50% higher in Alaska and Hawaii. Because the loan limit changes mechanically and only as a function of national housing prices, local housing supply or demand conditions have no effect on the jumbo loan cutoff. We exploit this fact in developing our instrument for housing price growth below.
Starting in the early 1980s, securitization moved beyond the GSEs, as private investment banks began to purchase and securitize jumbo loans, providing similar services for large mortgages that Fannie and Freddie provide for non-jumbos, although without the government subsidy. This fact can be seen by the jump in the average mortgage interest rate around the jumbo loan cutoff. In fact, this rate differential increased sharply during the financial crisis as
Both the moves to allow geographical expansion of banks within and across states, as well as the expansion of GSEs and private securitization have benefited both lenders and borrowers. Diversification opportunities have been enhanced, credit can flow more easily toward high-return investments, and opening up of markets has increased competition and lowered the price of credit (Demsetz and Strahan, 1997; Stiroh and Strahan, 2003). As we show next, however, financial integration has led to greater volatility, both in the housing sector (and, by extension, for the value of collateral more generally) and also in the economy as a whole.
III. FINANCIAL INTEGRATION AND HOUSE-PRICE VOLATILITYIn this section we test how financial integration affects the volatility of housing prices within local markets, and how the synchronicity (or interrelatedness) of housing price changes between markets varies with pair-wise measures of financial integration. In our first set of models, we build a panel dataset based on house-price volatility and financial integration at the level of the Central Business Statistical Area (CBSA) over the 1994 to 2006 period (unit of analysis = CBSA-year). In the second set of models, we build a richer panel by creating all CBSA-year pairs, again over the 1994 to 2006 period (unit of analysis = CBSA-pair-year). We test whether the correlation or similarity of housing prices shocks between pairs of markets changes as the two markets become more financially integrated with each other.
To start, we measure the volatility of the housing prices using the absolute deviation of housing price growth in a CBSA-year from the conditional mean, after removing time and CBSA
fixed effects. Specifically, we estimate the following regression:
Data for housing price growth rates are constructed from the Federal Housing Finance Association’s (FHFA) CBSA-level house price index. The residual growth-shocki,t captures how much housing prices growth differs in each CBSA and year compared to average housing price growth in this year across all geographies. The absolute value of this residual reflects housing price fluctuations specific to a given geography: Voli,t=|growth-shocki,t |.
The CBSA-year regressions test how integration affects housing-price volatility, as
where Integrationi,t equals our measures of the extent to which financial activity in a CBSA-year is connected to financial activity in other CBSAs (defined below).
The pair-wise regressions have the following structure:
Interrelatednessi,j,t = αt + γi,j + β2Integrationi,j,t + Other Controls + εi,j,t (3a)
where Interrelatednessi,j,t equals the negative of the absolute value of the difference in housingprice growth shocks between two CBSAs in a given year:
So, an increase in Interrelatednessi,j,t measures a decline in the difference in growth shocks between two CBSAs. In Equation (3a), Integrationi,j,t measures the pair-wise connectedness of two CBSA markets in a given year (defined below).
integrated lenders condition their credit decisions more on prices and less on other dimensions of credit risk (e.g. specialized knowledge about the local economy), or because capital flows more easily toward high-demand markets and away from low-demand markets. Both channels imply β1 0 in Equation (2). By looking at integration’s effects on pair-wise markets, we can isolate the capital flows channel. Imagine two CBSA markets – ‘A’ and ‘B’ – that are well integrated.
A shock to prices in ‘A’ (and thus to credit demand there) will draw financial resources away from ‘B’, thus accommodating the credit demand and raising prices in A and lowering them in B.
This second capital flight channel thus suggests that financial integration ought to make houseprice changes become less correlated as integration between two markets increases, so β2 0 in equation (3a).5 Measuring Financial Integration by CBSA-year Our measure of financial integration is built from the distribution and ownership of bank branches and deposits across local markets. The measure is based on information on total deposits, location and ownership of all bank branches from the Federal Deposit Insurance Corporation’s (FDIC) Summary of Deposits, available online annually from 1994 forward.6 We construct the In-CBSA ratio, equal to the fraction of all deposits in a CBSA that are owned by a holding company which also owns deposits in one or more other CBSAs.7
5 House price variation driven by local credit supply shocks will tend to attenuate this effect.
6 See http://www2.fdic.gov/sod/.
7 We define a banking company as the highest entity within a bank holding company for banks owned by holding companies, or for the bank itself for stand-alone banks.
which in turn may reflect risk management or diversification motivations of potential entrants.