In the aftermath of the 2007–09 financial crisis, central banks around the world have sought to stimulate the economy through new policies specifically designed to revamp credit and housing markets. These policies include the US Federal Reserve’s QE1 and QE3, the ECB’s Targeted Longer-Term Refinancing Operations, and the Bank of England’s Funding for Lending Scheme, among others.
A main goal of these unconventional policies was to make it cheaper for lenders to access funds and, in turn, “to enhance the functioning of the monetary policy transmission mechanism by supporting lending to the real economy”.1
Stimulating lending activities can be a powerful way to support the housing sector and foster consumer spending. However, the literature has identified several frictions in the mortgage market that could alter the transmission of monetary policy to credit markets and to the real economy. These include product design (Agarwal et al. 2017a, Agarwal et al. 2017b, Greenwald 2018); fixed- versus adjustable-rate contracts (Di Maggio et al. 2017); and lender market power (Scharfstein and Sunderam 2016, Xiao 2020).
In a recent paper (Benetton et al. 2021), we seek to advance our understanding of the effects of central bank policies on credit markets by studying the UK residential mortgage market around the introduction of the Bank of England’s Funding for Lending Scheme, a central bank facility that offers cheap medium-term loans to UK lenders.
Specifically, we examine a novel channel that affects the transmission of central bank policies to heterogeneous households via credit markets: lenders’ indirect price discrimination strategies through menus of two-part tariffs composed of origination fees and interest rates.
We combine different data sources to gain a broad picture of UK mortgage markets. Critically, we include loan-level data on the universe of residential mortgages originating around the onset of the Funding for Lending Scheme, as well as lenders’ drawings on Funding for Lending Scheme funds. These data allow us to describe some notable institutional features of the UK mortgage market, such as posted rates and fees equal across borrowers and mortgages with fixed interest rates for only a relatively short period (e.g. two years), which encourage borrowers to remortgage frequently and, as a result, generate a large fee income for lenders.
Moreover, these rich data allow us to provide new evidence on lenders’ market segmentation and pricing strategies, most notably their pervasive use of menus with two-part tariffs that combine (fixed) origination fees and interest rates. The pricing literature shows that indirect (i.e. second-degree) price discrimination through menus of two-part tariffs allow sellers to segment heterogeneous buyers and extract surplus from them (Wilson 1993).
In the mortgage market, lenders observe some of this heterogeneity, but they may not be able or do not want (e.g. Rotemberg 2011) to directly condition their prices on observable demographic characteristics such as income, age, or geographic region. However, this heterogeneity leads different borrowers to select different loan amounts, and thus menus of two-part tariffs effectively allow lenders to increase their profits.
New evidence of price discrimination in the mortgage market
We document three main empirical regularities. First, on average, lenders ask for an interest rate premium of 27 basis points on mortgage products with a £1,000 lower origination fee. Second, after the introduction of the Funding for Lending Scheme, which decreased their funding costs, lenders decreased interest rates but increased origination fees (Figure 1). Third, the number of interest rate/fee combinations for each product type has surged over time.
Figure 1 Mortgage pricing and the introduction of the Funding for Lending Scheme
Overall, these patterns suggest that lenders have been actively seeking to price-discriminate across borrowers using two-part tariffs, most notably after the introduction of the Funding for Lending Scheme to access cheaper funding. There is further suggestive evidence that borrowers may be paying more attention to interest rates than to fees in their mortgage choices.
We next seek to understand how borrowers choose among the available mortgage products, and how lenders set their rates and fees depending on their funding costs. To this end, we develop an equilibrium model of the mortgage market that incorporates the main features found in our descriptive analysis and estimate it using our rich datasets.
On the demand side, heterogeneous borrowers, who may have different sensitivities to interest rates and origination fees, make a discrete choice of the optimal mortgage product and a continuous choice of the optimal loan amount.
On the supply side, lenders offer differentiated mortgage products and maximise expected profits by setting interest rates and origination fees as well as by choosing to borrow (or not) from the central bank. Central bank policies affect lenders’ costs and, through them, lenders’ pricing.
Our demand estimates reveal that borrowers are significantly more sensitive to interest rates than to origination fees, and this is more pronounced among households with low income and with a young head.
The discrete-choice product demand is more elastic to interest rates than the continuous-choice loan demand: a 1% increase in the interest rate on a given product is associated with an average drop of about 7.5% in its market share, but with a fall of only 1% in its average loan size. This suggests that borrowers shop across lenders and products for a lower interest rate, focusing much less on origination fees.
Our supply estimates suggest that the Funding for Lending Scheme led to a reduction in lenders’ funding costs by approximately 70 basis points. Given an average marginal cost of approximately 350 basis points in the quarters before the introduction of the Scheme, the Funding for Lending Scheme decreases marginal costs by approximately 20%.
Using our equilibrium model, evaluated at the estimated parameters, we decompose the overall surplus increase due to the decrease in lenders’ funding costs through the scheme into lenders and borrowers. Our parameterised model implies that lenders decreased posted interest rates by approximately 40 basis points but increased posted fees by approximately £120.
These changes are consistent with our descriptive evidence of Figure 1, suggesting that our model includes the economic forces that account for them. More substantively, our model implies that the Funding for Lending Scheme boosted aggregate lending by more than 30%.
We also perform an extensive analysis across different demographic groups of different borrower outcomes and welfare to understand the welfare implications of the large heterogeneity across groups. We find that households in areas with higher house prices (and thus higher loan sizes), such as London and Southeast England, enjoyed the largest gains in consumer surplus due to the Funding for Lending Scheme because the associated decrease in interest rates favoured areas with the highest concentration of borrowers with larger loans.
Finally, we use our model to understand the contribution of indirect price discrimination through two-part tariffs to market outcomes and welfare. Specifically, we ban lenders from charging origination fees. In such a counterfactual scenario with zero origination fees, lenders charge higher interest rates to offset the drop in profits due to the ban on fees. Naturally, lenders’ profits decline but consumer surplus also decreases, indicating that they benefit from this market segmentation and price discrimination increases surplus.
In the context of our main research question, this finding demonstrates that two-part pricing provides a mechanism through which household heterogeneity amplifies the effects of monetary policy.
Implications for the transmission of monetary policy
Heterogeneity in the interest rate elasticity of loan demand, both across borrowers and relative to the origination fee elasticity, has important implications for the transmission of monetary policy. If consumers are more sensitive to changes in rates than to changes in fees, lenders could respond to a fall in their funding costs by offering larger interest rate reductions in ‘exchange’ of higher origination fees.
The lenders’ two-part pricing strategy represents a mechanism through which household heterogeneity amplifies the effects of monetary policy on lending. The reason is that menus of two-part prices allow lenders to offer lower rates to the most rate-sensitive households. While lenders profit from origination fees, in a world where fees are banned, most mortgagors – especially the more rate-sensitive ones such as younger and lower-income households – would borrow less because lenders would charge higher rates on average to offset the ban on fees.
Agarwal, S, I Ben-David and V Yao (2017a), “Systematic mistakes in the mortgage market and lack of financial sophistication”, Journal of Financial Economics 123(1): 42–58.
Agarwal, S, S Chomsisengphet, N Mahoney and J Stroebel (2017b), “Do banks pass through credit expansions to consumers who want to borrow?”, The Quarterly Journal of Economics 133(1): 129–90.
Benetton, M, A Gavazza and P Surico (2021), “Mortgage pricing and monetary policy”, CEPR Discussion Paper 16456.
Di Maggio, M, A Kermani, B J Keys, T Piskorski, R Ramcharan, A Seru and V Yao (2017), “Interest rate pass-through: mortgage rates, household consumption, and voluntary deleveraging”, American Economic Review 107(11): 3550–88.
Greenwald, D L (2018), “The mortgage credit channel of macroeconomic transmission”, mimeo, Massachusetts Institute of Technology.
Rotemberg, J (2011), “Fair pricing”, Journal of the European Economic Association 9(5): 952–81.
Scharfstein, D, and A Sunderam (2016), “Market power in mortgage lending and the transmission of monetary policy”, mimeo.
Wilson, R B (1993), Nonlinear pricing, Oxford University Press.
Xiao, K (2020), “Monetary transmission through shadow banks”, Review of Financial Studies 33(6): 2379–420.