Limits of Macroprudential Policy, and Impacts of Healthcare Costs on Small Firms
Job Market Paper Roundup
Last year I talked about two of my PhD students on the market — Abhinav Gupta (now at UNC) and Abhishek Bhardwaj (now Tulane). It’s job market season again, and I have a couple of other students whose research I want to share, along with some other job market paper roundups.
Dynamic Responses to Macroprudential Policy
One of my students is Iris Yao — in her job market paper, “Household Responses to Macroprudential Policy,” she studies one of the most important policy shifts since the 2008 financial crisis — the introduction of macroprudential policies intended to regulate household debt. These types of policies have been widely introduced in many countries in response to concerns about household overleverage, and the broader externalities and risks associated with these loans. Consequently, many countries have imposed limits on loan-to-value as well as debt-to-income ratios.
Prior literature in this area has taken these macroprudential rules as static — they apply at the time of loan origination, and it is assumed that households take these limits as given. As a result, the dynamic response of households to these debt limits has usually been ignored, despite the fact that whether households “undo” these leverage limits through subsequent borrowing is of central importance in evaluating the effectiveness of such policies.
Iris studies macroprudential changes in Canada using a full panel of consumer debt histories. Canada has seen large increases in household leverage for the past fifteen years, motivating tighter limits on debt-to-income ratios at the time of mortgage origination. These limits apply to the combination of mortgage debt, as well as non-mortgage debt (as in other countries).
One of Iris’ most striking findings is how households manipulate their non-mortgage debt around debt thresholds. Her results really surprised me — households lower their revolving consumer debt prior to mortgage origination in order to qualify for mortgages — and then re-lever back up right afterwards. This is a really fascinating piece of consumer credit behavior which shows up for mortgage originations in general in the Canadian setting: households are effectively gaming the system, with respect to non-mortgage debt, right around the time of mortgage applications! Looking at the macroprudential policy reform helps to identify that this pattern is due to mortgage limits specifically. Basically, when these limits tighten; households respond by cutting back revolving credit even more, before letting that revolving debt bounce back after mortgage origination. You can see this result below:
Given this pattern of dynamic household re-leveraging — a natural question to ask is whether consumer responses invalidate the effectiveness of macroprudential policy in general. Perhaps surprisingly, the answer is no: the macroprudential limit changes were, indeed, effective in lowering household mortgage debt, credit, and limiting house price growth — the intended purpose of the policy.
The ultimate reason for this is that leverage limits are binding for mortgage lending — where they can be verified at the time of debt limits — but not binding for revolving debt, which can more easily be changed by borrowers around the time of borrowing. To further support this channel, Iris breaks out the effects of macroprudential policy depending on whether the mortgage payment is binding (which can only be relaxed by reducing the mortgage size) or whether the total payment is binding (which can be relaxed by lowering, temporarily, non-mortgage credit). House prices are more sensitive to the mortgage payment dimension, suggesting that macroprudential policy only has persistent effects to the extent that its impact on consumer borrowing is durable.
Private Equity, Healthcare Costs, and Entrepreneurship
My other student is Samantha Zeller whose Job Market Paper “Effect of Increasing Healthcare Costs on Small Firms: Evidence from Private Equity Acquisitions of Hospitals” concerns a topic that is of immense importance: why is there a decline in entrepreneurial activity? Because startups are so important for innovation and employment growth, these trends are concerning for what they imply about economy-wide stagnation. Previous research has highlighted the roles of changing demographics or credit supply in impacting young companies, while Samantha identifies a novel component — rising healthcare costs — which impose unique burdens on small companies.
While we all know that healthcare costs are steadily rising, connecting these shifts to firm outcomes is no easy task. To do so, Samantha has gone through a really arduous data process — collecting information on health insurance premiums and linking these to firm-level outcomes through the US Census.
In order to gain identification on this important question, Samantha adds an additional component: Private Equity (PE) acquisitions of hospitals. Tong Liu has a nice JMP which finds that PE buyouts lead to higher hospital prices. Samantha goes one additional step to establish that PE acquisitions also raise health insurance premiums which are passed on to local firms when insurance markets are more competitive. This is not an obvious step, depending as it does on market concentration in the health insurance market.
Samantha next connects these health insurance cost shocks with small firm employment shares, startup rates, and firm-level churn. Looking at these outcomes requires the merge with the confidential Census information, and gives her a really unique ability to investigate these outcomes for these small and young firms. Her main approach compares firms in markets exposed to higher insurance premia resulting from PE acquisitions, against control firms which are unaffected through a difference-in-differences design. The main findings are that higher health insurance premia adversely impact young firms, reducing both entry and inducing more exits. The broad implication of this finding is that higher economy-wide health costs adversely impact young firms, and so likely have broader negative consequences for innovation and growth than previously considered.
No offense intended to anyone I didn’t include; just a few papers I’ve seen so far that seemed interesting to me:
Vrinda Mittal has a JMP arguing that underfunded public pension plans give money to low quality Private Equity funds, leading to capital misallocation and worse productivity for the firms they buy. She’s a co-author on my other work on real estate shifts associated with remote work.
Another co-author, Nishaad Rao, looks in his JMP at the long-run implications of booming labor markets of intergenerational wealth accumulation. Real estate returns capitalize the effects of good labor market outcomes, resulting in wealth gains which are passed down intergenerationally.
Roman Rivera’s JMP is about electronic bracelet monitoring. This is something that the (late) Mark Kleiman was really excited about as a way to avoid costly incapacitation. It turns out electronic monitoring does deter crime — but also results in additional violations of the monitoring.
Tom Cui finds that the post-war Great Migration of Black Americans to the North helped to spark exclusionary zoning practices.
Xuequan Elsie Peng argues the Bloomberg-era rezoning increased the city’s floorspace by 5% — better than nothing, but not nearly enough to meet the huge demand.
Fernando Cirelli at NYU Econ finds that inflation really makes it harder for low-income people to engage in precautionary/buffer savings, which they leave in bank deposits and gets eroded away under inflationary scenarios. It can sometimes be hard to find welfare consequences of low-moderate inflation, but this is a natural channel.
Finally, not a job market paper, but just a really neat paper by Clayton Nall, Christopher Elmendorf, and Stan Oklobdzija drives a nail in the coffin of the homevoter hypothesis. Households don’t oppose new local construction because they are trying to raise prices and preserve their own values. Instead, people don’t really seem to understand supply and demand dynamics for home construction; and just (mostly falsely) believe that new construction won’t lower prices. A positive spin here is that there is value in the YIMBY project — just explaining to people more construction will (at least when done a large enough scale) likely will lower prices.