Jed Kolko

The Effects of Mobile Phones and Hands-Free Laws On Traffic Fatalities

Using state-level panel data on mobile phone ownership, hands-free laws, and traffic fatalities, this paper finds that mobile phone ownership is associated with higher traffic fatalities, but only in bad weather or wet road conditions. The limited experience of a few states suggests that hands-free laws reduce traffic fatalities, but again only in bad weather or wet road conditions, and possibly also in rush-hour traffic. The findings suggest that the benefits of hands-free laws depend on driving conditions, and an important policy implication is hands-free laws should be more strongly enforced when driving conditions are more difficult. These findings on hands-free laws differ from other research, which consistently finds that hands-free and handheld phone usage have similar effects on driver distraction and accident risk, calling into question the usefulness of hands-free laws.

The Determinants of Gentrification

This paper assesses why lower-income urban neighborhoods gentrify. Over the period 1980-2000, gentrification was more likely in Census tracts that are closer to the city center and have older housing stock, consistent with theoretical predictions from classic urban models and with other recent empirical work on gentrification. The paper makes three contributions. First, neighboring tract income is shown to contribute to gentrification, providing evidence of positive inter-neighborhood spillovers. Second, the reasons for gentrification are shown to vary across cities: proximity to the city center and an older housing stock contribute more to tract-level gentrification in metropolitan areas where these characteristics are scarce - larger and newer metropolitan areas, respectively. Accordingly, U.S. regions vary in how well their cities fit the general pattern of gentrification: cities in the South and Midwest exhibited gentrification over the period 1990-2000, whereas gentrification was characteristic only of the Northeast over the period 1980-1990. Finally, gentrification is accompanied by increases in the number of households and a growing housing stock, as well as changes in residential demographic composition. The analysis relies on the Neighborhood Change Database, which provides tract-level decennial Census data recalculated using a consistent definition of tract boundaries.

Employment Location, Neighborhood Change, and Gentrification

This paper assesses the contribution of employment location to neighborhood change and to gentrification. At the tract level, average household income change from 1990 to 2000 is positively correlated both with the change and the level of average earnings for nearby jobs. I instrument for the tract-level change in average earnings with predicted change in tract average earnings based on national industry trends, confirming a causal interpretation. The relationship between employment location and neighborhood change is strongest for tracts with slower commuting speeds, suggesting that people live nearer work where the time cost of commuting is higher.

Is the ‘Shop Around the Corner’ a Luxury or a Nuisance? The Relationship between Income and Neighborhood Retail Patterns

Affluent neighborhoods present a potentially attractive location for retail establishments because of their higher purchasing power and demand for a wide range of specialized goods and services. However, if high income households perceive retail in general – or certain types of retail, such as Big Box stores – as an undesirable use, they may be able to block commercial development through zoning and the political process. In thispaper we shed light on these issues by examining the relationship betweenneighborhood income and several different types of retail presence for 58 large U.S metropolitan areas. We combine detailed data from the National Establishment TimeSeries database on retail establishments and employment, by industry category and firm type, with Census data on ZCTA income and demographics. Results indicate that retail density varies with income for certain retail types, such as food service and chain supermarkets and drugstores. In addition, average establishment size increases with income for all retail types. Retail density increases with population density, as expected, and decreases with distance to CBD and with share of owner-occupied housing.

Why Should Governments Support Broadband Adoption

Governments justify support of Internet diffusion on two grounds: (1) to overcome a persistent digital divide in broadband availability and (2) to facilitate online activities that are socially or economically desirable. This paper assesses both these claims. Using individual-level data from Forrester Research, the analysis finds significantly lower residential broadband adoption in lower-income and lower-density zip codes, controlling for individual characteristics. Further tests show that lower adoption in these areas is evidence of a persistent digital divide in availability. The analysis then assesses how broadband adoption changes individuals’ usage of online activities. Broadband adoption increases individuals’ frequency of researching health information online, but there is no evidence that broadband adoption increases usage of online job sites or online government services. Localities currently considering municipal wireless (Wi-Fi) initiatives should focus on digital divide justifications rather than expecting to raise usage of a wide range of online activities perceived to be socially desirable.

5. CONSUMERS AND CITIES

Urban economics has traditionally viewed cities as having advantages in production and disadvantages in consumption. We argue that the role of urban density in facilitating consumption is extremely important and understudied. As firms become more mobile, the success of cities hinges more and more on cities’ role as centers of consumption. Empirically, we find that high amenity cities have grown faster than low amenity cities. Urban rents have gone up faster than urban wages, suggesting that the demand for living in cities has risen for reasons beyond rising wages. The rise of reverse commuting suggests the same consumer city phenomena.

Employment Growth Around New Transit Stations in California

This paper assesses employment growth around more than 200 new transit stations in California that opened between 1992 and 2006. Using difference-in-differences estimation, employment growth within a quarter-mile of new stations was, on average, no faster after stations opened than before they opened, relative to comparison areas; the effects were heterogeneous across individual transit stations. The comparison areas are matched blockgroups, selected for their similarly to station areas on economic, demographic, and geographic variables.Employment changes are measured precisely using the National Establishment Time-Series (NETS) database, a longitudinal establishment-level dataset that includes employment counts and exact street addresses over time. The analysis overlays geocoded NETS employment data with geocoded transit stations and Census blockgroup-level data.

Moving Across Borders: The Determinants of Interstate Business Relocation

Businesses moving to other states spark political accusations and concerns about state business climates. However, previous work has demonstrated that interstate business relocation is rare. This short note documents the considerable variation in business relocation rates among states and tests several explanations for this variation. The same states tend to exhibit higher rates of relocation both into and out of the state. Relocation accounts for higher shares of gross employment changes in states where more economic activity is located near the state border. Net employment change has only a modest relationship with in-migration rates and no significant relationship with out-migration rates, confirming that jobs leaving the state are not a bellwether of poor state economic performance.