With housing debt accounting for over 80% of total UK household liabilities, choosing a mortgage is one of the most important financial decisions consumers make. It is also a very difficult one to get right. Even after deciding which house to buy and how much to borrow, a consumer is faced with comparing dozens, sometimes even hundreds, of mortgage products with many different features, such as the duration of the fixed-rate period, and multiple price components. It is hardly a surprise that regulators around the world, including the FCA, are interested in how well consumers can make decisions about their mortgages and what outcomes they experience. This paper adds to the evidence on these topics by investigating a uniquely granular combination of lending transaction reports, detailed product information and credit files for nearly 700,000 UK households that took out a mortgage between January 2015 and July 2016.
I also develop and implement a new methodology for measuring consumer outcomes in the mortgage market or other circumstances with multidimensional product prices and features. The proposed methodology - dominance analysis - focuses on choices of unambiguously ‘dominated’ products by assessing, for each observed borrower, whether they were eligible for another product with the same features as their chosen mortgage but at a strictly lower price (i.e. having rates or fees that are lower with none that are higher). This makes it possible to detect consumers who are likely to not be searching the market effectively without making trade-offs between different price components or additional assumptions about consumer preferences for different mortgage types.
I find that nearly one in three UK consumers (29.9%) in 2015-16 chose mortgage products that were dominated by apparently available alternatives and paid £550 more per year on average as a result. Lower incomes, worse credit scores, higher age and facing a more time-constrained or complex decision are associated with higher rates of dominated choices. Evidence suggests that limited transparency about eligibility for different products and borrowers’ tendency to choose lenders with whom they have an existing relationship are also likely to be among the contributing factors