Volume 40, Issue 2 (2022)

Opportunity Zones: Gambling with Our General Welfare

Cedar Riverside Opportunity Center in Minneapolis, MN.

Jen L. Davison*

The economic inequality gap in America is in a crescendo. From WWII to the 1970s, economic growth was generally steady and prosperity was increasingly shared, but since then growth has slowed, the income gap has increased, and wealth has become highly concentrated among fewer people.[1] An alarming 45 million Americans now live below the poverty threshold,[2] with 140 million classified as poor or low wealth.[3] The distribution of this remarkable economic inequality is not evenly spread. For the past forty years, “[mi]xed-income neighborhoods, and with them mixed-income schools and playgrounds, have been replaced by a rapidly growing number of neighborhoods that are either very poor or very affluent.”[4]

When the market fails neighborhoods, legislation can intervene. The Constitution creates a fundamental tie between federal taxing and spending powers and the general welfare of America’s people.[5] Article I of the Constitution spells out this power-bound-by-intent relationship: “The Congress shall have Power To lay and collect Taxes, Duties, Imposts and Excises, to pay the Debts and provide for the . . . general Welfare of the United States.”[6] During the Great Depression, the Supreme Court definitively adopted the perspective that taxing and spending are constitutionally constrained by their intent, which must be for the general welfare.[7] Fiscal policies based in legislation flowing from this authority have the advantage of being able to “target specific communities.”[8] Out of these powers come federal tax bills, including the Tax Cuts and Job Acts (TCJA) and its creation of Opportunity Zones (OZs).

Opportunity Zones Create Particular—Not General—Welfare

Sean Parker was an early investor in Facebook who was worried about how much he would owe in taxes when he sold his stock.[9] Joining forces with other billionaires, his coalition began lobbying Congress for a new tool to deal with their capital gains.[10] As a result, at the end of 2017, the TCJA was passed, heralding the dawn of a seemingly uncontroversial new program called Opportunity Zones.[11] The program was built to limit the capital gains tax liability of corporations and financiers while spurring the economic revitalization of America’s poorest neighborhoods.[12]

TCJA empowered each governor to designate one quarter of their low-income, high-poverty census tracts in their state or territory—often marked by high unemployment, population decline, vacant buildings, and residents with low income and/or education levels—as OZs.[13] Meanwhile, hedge funds and banks created Qualified Opportunity Funds (QOFs), investment vehicles organized to hold 90% or more of their assets in OZs.[14] Investors could qualify for the program if they located 70% of their tangible property in an OZ.[15] Once investors rolled their unrealized capital gains into a QOF, they began temporarily deferring some of their capital gains taxes.[16] The longer they retained that investment, the greater the guaranteed tax relief.[17] However, there were no such guarantees of economic benefit for the communities identified for investment.[18] Their benefit from OZs was a gamble. 

OZ Reforms and the General Welfare

So far, OZ investors can look forward to cashing in on tax-free capital gains.[19] In contrast, economically impoverished communities can look forward to mostly unrealized gains.[20] These results suggest that Congress has delegated its spending power by tax incentive in the OZ program, but it has not effectively delegated the general welfare intent that serves to guide and limit that spending power. Among his economic plans, President-elect Biden has proposed reforms to the OZ program.[21] Perhaps these reforms will be the necessary tether binding OZs to the welfare of our economically depressed communities; those reforms are sorely needed.[22] But the tale of OZs also serve as a reminder. In a time when private-government partnerships are an increasingly new normal for the U.S.,[23] and when COVID-19 is causing widespread economic desperation,[24] Congress should enact fiscal policies that do not surrender general welfare outcomes into the hands of those driven by particular welfare outcomes.

 


 

* J.D. Candidate (2021), University of Minnesota Law School. The author lives in an Opportunity Zone in Cedar-Riverside, Minneapolis, Minnesota.

[1] Chad Stone et al., A Guide to Statistics on Historical Trends in Income Inequality, Ctr. on Budget & Pol’y Priorities, (Jan. 13, 2020), https://www.cbpp.org/research/poverty-and-inequality/a-guide-to-statistics-on-historical-trends-in-income-inequality.

[2] IMF, United States of America: Staff Concluding Statement of the 2019 Article IV Mission (2019), https://www.imf.org/en/News/Articles/2019/06/06/mcs060619-united-states-staff-concluding-statement-of-the-2019-article-iv-mission.

[3] Reverend William Barber, Opinion, Trump’s Greatest Vulnerability is the Economy—Just Ask Poor Americans, The Guardian (Feb. 11, 2020), https://www.theguardian.com/commentisfree/2020/feb/11/trump-economy-poor-americans-vulnerability; The Daily Social Distancing Show, Thomas Piketty—Why Capitalism Must Be Reformed, YouTube (May 6, 2020), https://www.youtube.com/watch?v=XtahuPcsNVM (featuring economist Thomas Piketty who observes, “In the past three decades, we’ve seen a lot more billionaires, but we’ve seen a lot less growth. And so in the end, the idea that you get prosperity out of inequality just didn’t work out.”).

[4] Sean F. Reardon & Kendra Bischoff, No Neighborhood is an Island, in The Dream Revisited: Contemporary Debates About Housing, Segregation, and Opportunity in the Twenty-First Century 56 (Ingrid Gould Ellen & Justin Peter Steil eds., 2019).

[5] Troy Segal, Monetary Policy vs. Fiscal Policy: What’s the Difference?, Investopedia.com (May 5, 2020), https://www.investopedia.com/ask/answers/100314/whats-difference-between-monetary-policy-and-fiscal-policy.asp.

[6] U.S. Const. art. I, § 8, cl. 1.

[7] United States v. Butler, 297 U.S. 1, 64 (1936) (“The true construction undoubtedly is that the only thing granted is the power to tax for the purpose of providing funds for payment of the nation’s debts and making provision for the general welfare.”); Helvering v. Davis, 301 U.S. 619, 641 (1937):

Nor is the concept of the general welfare static. Needs that were narrow or parochial a century ago may be interwoven in our day with the well-being of the nation. What is critical or urgent changes with the times. The purge of nation-wide calamity that began in 1929 has taught us many lessons. Not the least is the solidarity of interests that may once have seemed to be divided. . . . Rescue becomes necessary irrespective of the cause. The hope behind this statute is to save men and women from the rigors of the poor house as well as from the haunting fear that such a lot awaits them when journey’s end is near.

In deciding if a particular expenditure is intended to serve the general welfare, courts defer to Congress “unless the choice is clearly wrong, a display of arbitrary power, [or] not an exercise of judgment.” Helvering, 301 U.S. at 640.

[8] See Segal, supra note 6.

[9] Jesse Drucker & Eric Lipton, How a Trump Tax Break to Help Poor Communities Became a Windfall for the Rich, The Bonanza, N.Y. Times (Aug. 31, 2019), https://www.nytimes.com/2019/08/31/business/tax-opportunity-zones.html (quoting Parker who said, “When you are a founder of Facebook, and you own a lot of stock, . . . you spend a lot of time thinking about capital gains”).

[10] Ctr. for Responsive Politics, Client Profile: Economic Innovation Group, https://www.opensecrets.org/federal-lobbying/clients/summary?cycle=2015&id=D000068054,

[11] Tax Cuts & Jobs Act, H.R. 1, 115th Cong., § 13823 (2017), https://www.congress.gov/115/bills/hr1/BILLS-115hr1enr.pdf. Six pages of the Act are devoted to Opportunity Zones and are now housed in the Internal Revenue Code under Title 26, Subtitle A-Income Taxes, Chapter 1-Normal Taxes and Surtaxes, Subchapter Z-Opportunity Zones. This OZ program was never debated on the floor of the House or Senate. See Jim Tankersley, Tucked Into the Tax Bill, a Plan to Help Distressed America, N.Y. Times (Jan. 29, 2018), https://www.nytimes.com/2018/01/29/business/tax-bill-economic-recovery-opportunity-zones.html.

[12] See Tankersley, supra note 11.

[13] Treas. Reg. § 1400Z-1 (2019).

[14] Treas. Reg. § 1400Z-2(a) (2019).

[15] When combined, the 90% QOF requirement and the 70% tangible property requirement means QOFs set a floor of 63% investment in an OZ. Alec Fornwalt & John Buhl, Treasury Department Proposes New Regulations for Opportunity Zones, Tax Found. (Oct. 22, 2018), https://taxfoundation.org/new-proposed-opportunity-zones-regulations/.

[16] Treas. Reg. § 1400Z-2(a)-(b) (2019).

[17] Treas. Reg. §§ 1400Z-2(b)(B)(iii)-(iv), 1400Z-2(c) (2019). See also Forrest David Milder, INSIGHT: A Dozen Things You Should Know About Setting Up an Opportunity Fund, Bloomberg Tax (Nov. 29, 2018), https://news.bloombergtax.com/daily-tax-report/insight-a-dozen-things-you-should-know-about-setting-up-an-opportunity-fund?utm_source=twitter&utm_medium=taxdesk&utm_campaign=6pm.

[18] See Steven Bertoni, An Unlikely Group of Billionaires and Politicians has Created the Most Unbelievable Tax Break Ever, Forbes (July 18, 2018, 6:00 AM), https://www.forbes.com/sites/forbesdigitalcovers/2018/07/17/an-unlikely-group-of-billionaires-and-politicians-has-created-the-most-unbelievable-tax-break-ever/#2bbe65171485 (“If everything goes right, a big slice of the estimated $6.1 trillion of paper profits currently resting on American balance sheets is about to go to work to revitalize America’s depressed communities. If all goes wrong, however, it will prove to be one of the biggest tax giveaways in American history, all in service of gentrifying neighborhoods and expelling local residents.”)

[19] Theodore Schliefer, The New Hotness for Tech Billionaires? Do-gooder Investments They Can Write Off on Their Taxes, Vox (Oct. 16, 2018, 5:50 PM), https://www.vox.com/2018/10/16/17940120/opportunity-zones-sean-parker-silicon-valley-wealth-taxes (“[T]he whole point of tax breaks is that it incentivizes selfish behavior that the government thinks is good for society.”).

[20] Unrealized Gains: Opportunity Zones and Small Businesses, Smart Growth Am. & Democracy at Work Inst. (Oct. 2020), https://smartgrowthamerica.org/resources/unrealized-gains/.

[21] The Biden Plan to Build Back Better by Advancing Racial Equity Across the American Economy, “Reform Opportunity Zones to Ensure They Serve Black and Brown Communities, Small Businesses, and Homeowners” (2020), https://joebiden.com/racial-economic-equity/#.

[22] Nick Routley, Mapping the Uneven Recovery of America’s Small Businesses, World Econ. Forum (Oct. 6, 2020), https://www.weforum.org/agenda/2020/10/mapped-uneven-recovery-us-america-small-businesses-closure/.

[23] Jody Freeman, Private Parties, Public Functions and the New Administrative Law, 52 Admin. L. Rev. 813, 816, 818 (2000) (“Contemporary regulation might be best described as a regime of “mixed administration” in which private actors and government share regulatory roles . . . . Unlike agencies, however, [private actors] are not generally expected to serve the public interest.”).

[24] Lauren Bauer, Kristen E. Broady, Wendy Edelberg, & Jimmy O’Donnell, Ten Facts about COVID-19 and the U.S. Economy, Brookings Inst. (Sept. 17, 2020), https://www.brookings.edu/research/ten-facts-about-covid-19-and-the-u-s-economy/ (documenting facts ranging from decreased small business revenue to unemployment, and from rent nonpayment to food insecurity).

Does the Minneapolis Police Department Traffic Stop Data Reveal Racial Bias?

Does the Minneapolis Police Department Traffic Stop Data Reveal Racial Bias?[1]

Disclaimer: The views and opinions expressed in this article are those of the authors.

Introduction

George Floyd’s death at the hands of Minneapolis police reignited the national debate about racial discrimination by police and led to protests across the country.  Unfortunately, data on police behavior is limited; but one of the pieces of data that the Minneapolis Police Department (“MPD”) makes available can be used to determine if racial bias influences MPD traffic stops.  The analysis can be used to inform stakeholders and can be continually monitored to evaluate ongoing department practices.

Summary of Results

This study analyzed Minneapolis Police Department traffic stop data from 2016 to 2020 to determine if racial bias influences MPD behavior.  The Veil-of-Darkness method was used to conduct the analysis and was constructed to ensure that contextual and demographic factors are addressed.  Results of the analysis showed that Black drivers are 10.8% percent more likely to be stopped during the day, when officers can observe the driver’s race for profiling, than when Black drivers’ race is not observable during darkness. The effect was highly statistically significant and demonstrated that Minneapolis Police Department traffic stops are racially biased.

Data

Data was obtained from the Minneapolis Police Department website[2] logging 79,000 traffic law enforcements from October 31, 2016 to June 4, 2020.  Of those stops, approximately 69,000 have race captured.  Below is a table showing the distribution of traffic stops by driver race.  The underlying data captures Black and East African as separate races.  These groups were combined for the summaries in Tables 1 & 2 to facilitate comparison to U.S. Census, but separated for the statistical analysis as to align with MPD data reporting practices.

 

Data Source: Minneapolis Police Department[2]

 

Casual data analysis would conclude that police make racially biased decisions, because the data shows Black drivers (53.1%) are stopped much more frequently than White drivers (34.6%).  While the data is suggestive, it is critical to account for demographics and other factors, such as commuting behavior and geography.

Benchmarking is the technique most frequently used to account for demographic and other factors, such as commuting patterns and driving behavior in traffic stop data (Tillyer, Engel, & Cherkauskas, 2009).  This technique compares the frequency of traffic stops to the demographics of the geographic area.  Below is a comparison of the racial distribution of traffic stops from Table 1 to racial make-up of Minneapolis.  Demographic data for the City of Minneapolis comes from the U.S. Census Bureau and is current as of July 1, 2019 (United States Census Bureau, 2020). Note that the columns do not total to 100% because the “Other” segment has been omitted for the sake of comparison.

 

* Data Source: Minneapolis Police Department[2] & US Census Bureau [3]

 

This analysis demonstrates that Black and East African residents made up 19.4% of the Minneapolis population, but accounted for 53.1% of the City’s traffic stops.  Traffic stops with a White driver accounted for 34.6% of Minneapolis’ traffic stops, despite White residents making up 59.8% of the City’s population.  These results are consistent with a previously published study by the Hennepin County Special Litigation Office (Jany, 2018).  Comparison of traffic stops to the City’s demographics helps to overcome some of the previously noted shortcomings and provides some evidence of racial bias in the City’s traffic stop practices but, as stakeholders often note, other contextual factors, as mentioned above, may still not be fully accounted for.

 

Analysis

To analyze the traffic stop data without the need for benchmarking, economists Grogger and Ridgeway developed an alternative analytic method named the Veil-of-Darkness (Grogger & Ridgeway, 2006).  The Veil-of-Darkness hypothesis states that officers stop a larger share of minority drivers during the daylight when it is easier to see the drivers race for profiling purposes than they do during the night.  Limiting the analysis to the part of the day called the inter-twilight period addresses contextual factors such as commuting behaviors and geography.  Different commuting behaviors, such as rush hour traffic, might confound results where police presence or certain offenses are less (more) likely to occur, e.g. speeding, than non-rush hour traffic.  Because sunset varies throughout the year, it randomizes any geographic effects ensuring they do not influence results.  Together, the method gets to a controlled set of drivers based on driving habits across Minneapolis with the only varying factor being the presence of daylight.

The inter-twilight period is between 5:05p.m. and 9:41p.m. The times relate to the day in December when it gets dark the earliest and the day in June when the sun sets latest.  Below is a graph showing all traffic stops made by Minneapolis police from 10/31/2016 to 6/4/2020 with each dot representing a single traffic stop.  Traffic stops highlighted in red occurred during the inter-twilight period and are the focus of the analysis.

 

* Data Source: Minneapolis Police Department[2]

 

The Veil-of-Darkness method creates a natural experiment using the setting of the sun – in daylight police can more easily identify a car driver’s race than they can at night.  If the data shows that a larger share of Black drivers are stopped during daylight than at night, there is evidence of racial bias in the practices of the police.  Because the time at which the sun sets differs during the year, factors such as commuting behavior, police presence, and underlying criminal activity are effectively randomized in the sample and do not impact the results.  As a result, any difference identified by the analysis is exclusively due to the officer’s ability to perceive race using available daylight.

Below is a graph showing the share of Black drivers pulled over across thirty-minute segments of time during the inter-twilight period.

* Data Source: Minneapolis Police Department[2] and authors’ calculations

 

The chart shows the share of all traffic stops that include a Black driver during daylight and darkness.  This is done by grouping all traffic stops into half-hour windows.  The first set of data points, labeled 5:30 p.m. on the horizontal axis, show that 51.2% of traffic stops during the daylight included Black drivers while 49.4% of traffic stops during darkness involved a Black driver. Differing sunset times throughout the year make this comparison possible because sometimes during the year it is dark at 5:30 p.m. and other times it is light at 5:30 p.m.  Further analysis shows that Black drivers are stopped more frequently during daylight in nine of the ten timeframes.  The comparison reveals that Black drivers are consistently stopped more frequently during daylight when officers can see the car driver’s race than in darkness.  This visual is an important first step in the Veil-of-Darkness method but leaves two unanswered questions:

1) How large is the disparity?

2) Is the disparity statistically significant?

Grogger & Ridgeway and subsequent researchers have used regression modeling to both quantify the difference in stop rates and determine statistical significance.  As discussed previously, the Veil-of-Darkness method ensures that demographic and contextual factors are adequately addressed and only three pieces of data are required to complete the analysis – time of the stop, date of the stop, and the driver’s race.  In addition to the primary analysis, sensitivity tests are an important component in the process.  Sensitivity tests are done to ensure that small changes to model specification do not have a large impact on the model’s results.  Researchers including (Grogger & Ridgeway, 2006), (Ritter & Bael, 2009) have done this by adding a third time factor to the regression equation.  Results of the regression modeling and sensitivity testing are provided below in Table 3.  The figures below reflect the increase in traffic stop rates for Black drivers between darkness and daylight – positive numbers reflect an increase in traffic stops during daylight.

Using MPD data from 2016 to 2020, Black drivers are 10.8% percent more likely to be stopped during the day when officers can observe the driver’s race for profiling than when the Black driver’s race is not observable during darkness.  As noted, the effect is highly statistically significant with a 2% probability or less of occurring randomly.  Sensitivity tests using step & spline functions for time demonstrate that the results do not materially change from specification changes, further enhancing confidence in the results.

Conclusion

Because of its ability to address contextual, demographic, and other factors, as well as the low data requirements (only time of the traffic stop, date of the traffic stop, and driver’s race are needed), the Veil-of-Darkness method is becoming a go-to method for traffic stop research.  Results of this study are largely in line with a 2009 study that used data from the Minneapolis Police Department and the Veil-of-Darkness method (Ritter & Bael, 2009).  That study used traffic stop data from 2002, concluding that Black drivers were 5.5% – 7.0% more likely to be pulled over during daylight than at night.

Together, the results of this study and previous studies demonstrate racial bias against Black people in the traffic stop practices of the Minneapolis Police Department.  This study demonstrates that racial bias against Black drivers exists, based on the most recent data (including 2020); when the results are compared to similar previous studies, the racial bias appears to have increased in recent periods.

References

Jeffrey Grogger & Greg Ridgeway, Testing for Racial Profiling in Traffic Stops From Behind a Veil of Darkness, 101 J. of Am. Stat. Ass’n 878-87 (2006).

Joseph A. Ritter & David Bael, Detecting Racial Profiling in Minneapolis Traffic Stops: A New Approach, Cura Reporter 11-17 (2009).

Libor Jany, Hennepin County report finds stark racial disparities in traffic stops, Star Tribune (Oct. 5, 2018), https://www.startribune.com/hennepin-county-report-finds-stark-racial-disparities-in-traffic-stops/495324581/.

Rob Tillyer, Robin S. Engel, & Jennifer Calnon Cherkauskas, Best Practices in Vehicle Stop Data Collection and Analysis, 33(1) Policing: An International Journal 69-92 (2010).

U.S. Census Bureau: Quick Facts, U.S. Census Bureau, http://www.census.gov/quickfacts/fact/table/minneapoliscityminnesota,US/PST045219 (last visited June 10, 2020).

 

[1] Joseph Schneider, D.B.A. recently completed his doctoral studies specializing in behavioral economics.  His research has focused on asset pricing and criminal justice.

Alfred Ndungu, Ph.D. is a statistician whose previous research has focused on statistical theory and healthcare.

[2] http://opendata.minneapolismn.gov/datasets/police-stop-data/data

[3] https://www.census.gov/quickfacts/fact/table/minneapoliscityminnesota,US/PST045219