Policy Uncertainty and Bank Mortgage Credit

BIS Working Papers  |  No 820  | 
16 October 2019


The uncertainties associated with possible changes in government leadership or policy can affect the behaviour of firms through various channels, such as industry regulation, monetary and trade policy, and taxation. Indeed, a growing literature shows that non-financial firms cut back investment expenditure when they face policy uncertainty around elections. However, it is an open question how policy uncertainty would affect banks' lending behaviour. And it is an important one because financial institutions, which operate in a heavily regulated industry, are likely to face more uncertainty than non-financial firms do when the political landscape changes, and their response to such changes may have a ripple effect on the economy because of their role as intermediaries. 


It is empirically challenging to establish a causal relationship between policy uncertainty and banks' mortgage lending decisions. First, any observable change in bank lending is an equilibrium outcome reflecting both the credit supply from banks and demand from borrowers. Second, a relationship between uncertainty and banks' investment decisions can be endogenous, as an economic downturn can itself generate a great deal of political uncertainty. Thus, establishing a causal relationship requires an exogenous measure of political uncertainty. This paper attempts to address these challenges in two ways. First, we take advantage of high-frequency, geographically granular US loan-level data - confidential Home Mortgage Disclosure Act data - to control for changing local demand for loans. Second, we employ a plausibly exogenous measure of policy uncertainty: the timing of US gubernatorial elections. 


We show that banks reduce the supply of mortgage credit in the quarters before their headquarter states hold gubernatorial elections. In our baseline regressions, we focus on the types of loan that we consider relatively more costly to reverse - jumbo loans held on bank balance sheets - as models of investment under uncertainty suggest that irreversibility increases the information value of waiting to invest, causing investment to vary negatively with fluctuations in policy uncertainty over time. However, the results hold for a broader universe of loans, although the magnitude is weaker than for jumbo loans held on bank balance sheets. Banks reduce lending both in their headquarter states and other states, suggesting that the pattern in the data is unlikely to be driven by changing demand in banks' headquarter states. We document heterogeneity across banks in their sensitivity to electoral uncertainty. First, state-chartered banks are more sensitive to changes in their state's political leadership following elections, implying that potential changes to state bank regulations create an additional layer of uncertainty for state banks as compared with national banks. Second, riskier banks cut mortgage credit a little more than less risky banks. The mortgage lending cycle around elections is more pronounced for less certain elections, as measured by closely contested election races and term-limited elections, where the incumbent governor cannot run for a re-election due to a term limit.


We show that banks reduce the supply of jumbo mortgage loans when policy uncertainty increases, as measured by the timing of US gubernatorial elections in banks' headquarter states. We use high-frequency, geographically granular loan-level data to address an identification problem arising from the changing demand for loans: (1) The data allow for a difference-in-difference specification and for state/time (quarter) fixed effects; (2) we observe banks reduce lending not just in their home states but also outside their home states when their home states hold elections; (3) we observe important cross-sectional differences in the way banks with different characteristics respond to policy uncertainty. Overall, the findings suggest that policy uncertainty has a real effect on residential housing markets through banks' credit supply decisions and that it can spill over across states through lending by banks serving multiple states.

JEL codes: G21, G28

Keywords: Bank Mortgage Credit, Housing Market, Policy Uncertainty, Gubernatorial Elections