Going Underground: The impact of deterrence on the landlords' decision to operate informally

ABSTRACT

This paper examines how enforcement of regulations affects firms’ decision to operate formally, underground, or exit the market. The firms in our study are landlords of rental properties. Using a unique data set that identifies underground and exited properties, we find that enforcement makes going underground more likely when only fines are issued but less likely when enforcement requires abatement. Additionally, observing higher levels of enforcement among neighbouring properties (general deterrence) increases the likelihood of going underground when abatement is required. These findings are consistent with the predictions of a theoretical framework we develop. Accordingly, the study makes two contributions to the informality literature. First, that whether enforcement has a positive or negative impact on operating underground depends on whether that enforcement requires abatement. Second, it identifies both the extensive margin between informality and formality, and the margin between informality and exiting an industry.

KEYWORDS:

• Informal economy
• rental housing
• deterrence
• licencing
• housing regulation

JEL CLASSIFICATION:

• K25
• L51
• R38
• R52

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ As Kanbur (2009a) notes, firms that are licenced (formal) but do not fully comply fall into a ‘grey’ area between the formal and informal sector. Analogously, Ulyssea (2020) observes that firms can comply with regulations on the extensive margin by choosing whether to become formal, and on the intensive margin by deciding to what extent they should comply with regulations after becoming formal.

² As of 2019 Baltimore has moved towards private inspectors as part of their licencing program.

³ Two unit properties with an additional commercial unit are considered MFDs.

⁴ Since this measure has changed through the years, we have taken the 2014 value for all periods.

⁵ We establish a property owner as professional if the owner’s name includes, INC, LLC, TRUST, etc.

⁶ We note that in each of the specifications the $R^2$ is limited and there are likely substantial factors that go into landlords’ decision process to operate underground ins unobservable, such as their own finances and their private knowledge of the condition of the property. Given the low, $R^2$ we run a link test (Tukey 2016) to ensure goodness of fit. Specifically, we use the predicted values from each model and the predicted values squared in an auxiliary model to predict landlords’ decision to obtain a licence. As appropriate for a properly specified model, the predicted values are significant with a t-statistic of 38.93 for Column 1. The squared predicted value should not be significant in our auxiliary model if there are no additional-omitted variables and our model is property specified. We confirm the model is correctly specified, given that t-static on the squared of the predicted values is 0.01..

⁷ Registration fees were $35 per dwelling unit and $100 per structure plus $30 per dwelling unit for vacant structures.

Additional information

Funding

The work was supported by the Loyola University Maryland - Impact Fellowship .