USC Marshall

We examine human capital's role in the gap in the quality of financial services in high- versus low-income neighborhoods. Using a near-comprehensive panel of over 350,000 U.S. mortgage loan officers, we document large and persistent differences in
loan ocer productivity and performance. We nd that rms' hiring, promotion,and retention policies disproportionately assign inexperienced and low-performing employees to branches serving low-income customers. These \bad apples" are more likely to commit misconduct, and to write bad loans that borrowers cannot afford. These labor market factors are important in explaining the observed income-based disparities in the quality of financial services.

Prior research indicates experience effects in real estate: sellers (landlords) having bought when aggregate prices were high set elevated asking prices (rents), even decades later. We study the
propagation of such valuation errors through local price competition. House sellers (landlords) appear influenced not only by their own historical experiences, but also the experiences of nearby sellers (landlords). Competing with a “bust-acquired” house bought during 2008-2012 is associated with discounts exceeding 1%. Although bust-acquired houses comprise < 10% of all sales, the amplification of their price distortions appears sufficient to impact aggregated (e.g., zip code level) prices and rents.

We find that a corrupt local environment amplifies the effects of financial distress. Following regional spikes in financial misconduct, credit becomes both more expensive and harder to obtain for nearby borrowers -- even those not implicated themselves. This is particularly harmful for cash-constrained firms, which cut investment more sharply and lay off more workers during industry downturns. Moreover, we find that local waves of financial misconduct are a risk factor for bankruptcy.

Landlords appear to use stale information when setting rents. Among over 43,000 California rental houses in 2018–2019, those last purchased during 2005–2007 (the peak) rent for 2–3% more than those purchased during 2008–2010 (bust). Neither house nor landlord characteristics explain this “peak-bust rental spread.” To clarify the mechanism, we test cross-sectional predictions from a simple theory of rent-setting. We find empirical support for both reference dependence and distorted beliefs. In the first, monthly payments establish (recurring) reference points, against which gains or losses are measured. In the second, past sales prices distort landlords’ current estimates of house values/rents.

City level differences in industry-adjusted Tobin’s q, an estimate of the value created for shareholders, are large, and have widened sharply over the last twenty years. Proxies for a city’s appeal to high-skill workers, such as existing education rates and favorable weather, are strongly associated with Tobin’s q, both in levels and changes. These results indicate that shareholders have recently captured a bigger part of the benefits associated with superior locations. The higher stock prices of firms in these locations appear to be driven by future growth opportunities, rather than improvements in current operating efficiency.