Essay

The Labor Economics Case for the Diversity Visa Lottery

Patrick Kennedy *

Introduction

Congress faces tough choices in the coming weeks and months about the design of our immigration system. Among those choices: the fate of the Diversity Visa (DV) lottery program, which grants up to 50,000 visas each year to emigrants from countries that do not traditionally send many people to the United States. President Trump has placed the DV program squarely in his crosshairs, claiming the lottery allows other countries to send “their worst people.” 1 Rafael Bernal, What Is the Diversity Visa Lottery?, The Hill (Jan. 28, 2018, 7:30 AM EST), https://perma.cc/EE9D-ZB35. Furthermore, two of the five immigration proposals that were before Congress in recent months, but have not been acted upon lately, would eliminate the DV program entirely. 2 See Lisa Desjardins, Every Immigration Proposal in One Chart, PBS NewsHour (updated Feb. 13, 2018, 6:17 PM EDT), https://perma.cc/C7RF-7UUA . The Goodlatte bill would eliminate the lottery without replacing it, in effect reducing the number of immigrants that come from low-admission areas every year by at least 50,000. 3 See id. The Grassley/Cotton bill would also eliminate the lottery but replace it with increases in family-based and employment-based visas. 4 While it is tempting to assume this would not restrict the number of legal immigrants eligible for visas, the bill also restricts who is eligible for a family-based visa. See id.

Authors writing in many major political publications have spilled ink 5 See, e.g., Abed A. Ayoub, In Targeting Diversity Visa Program, Trump Gets it Wrong Again, The Hill (Nov. 2, 2017, 1:20 PM EDT), https://perma.cc/LGU8-4JQZ; Dara Lind & P.R. Lockhart, The Diversity Visa Trump Hates, Explained, Vox (updated Dec. 15, 2017, 1:29 PM EST), https://perma.cc/4E53-96HY. describing the diplomatic benefits and positive self-selection spurred 6 Immigration produces a sorting function, in that people most likely to choose to migrate are those with high levels of unobservable motivation or similar characteristics, making them the most likely to succeed in business. See Barry R. Chiswick, Are Immigrants Favorably Self-Selected? An Economic Analysis 1, 2-7 (IZA Discussion Paper No. 131, 2000) (“[E]conomic migrants tend to be favorably ‘self-selected’ for labor market success. That is, economic migrants are described as tending, on average, to be more able, ambitious, aggressive, entrepreneurial, or otherwise more favorably selected than similar individuals who choose to remain in their place of origin.”). by immigration programs like the DV, but there is a more fundamental economic case for keeping the DV lottery in place: Immigrants move to places where other immigrants live, so establishing a mechanism to encourage immigration from countries with few immigrants is critical. Without the DV program, America will lose an important advantage in the global war for talent over the long run.

I. The Diversity Lottery: What Is It and Who Benefits?

The Immigration Act of 1990 established the DV lottery system, 7 Pub. L. 101-694, §§ 131, 104 Stat. 4978, 4997-99 (codified as amended at 8 U.S.C. § 1153(c) (2016)). which grants lawful permanent residence visas to up to 50,000 applicants per year from “low-admission” countries. 8 See 8 U.S.C. § 1153(c)(1)(B)-(E) (2016). Low-admission countries must have sent fewer than 50,000 immigrants to the U.S. in the five preceding fiscal years. 9 Id. § 1153(c)(1)(A)-(B). This is updated annually based on the previous five fiscal years. Countries like Mexico, Canada, India, and China thus do not qualify for the DV lottery in a typical year. 10 See, e.g., Visa Bulletin for August 2018, U.S. Dep’t State (July 10, 2018), https://perma.cc/M9DX-NJP3. Instead, the program is targeted toward countries such as New Zealand, Angola, Mongolia, Libya, and many others. 11 See U.S. Dep’t of State, Instructions for the 2019 Diversity Immigrant Visa Program (DV-2019) 16-18 (2017), https://perma.cc/PA4C-ZMRJ. Prospective immigrants apply online or at an embassy, with a hard deadline set in advance each year. There is no cap on the number of applicants. Lottery numbers are then pulled from the applications submitted before the deadline. Unlike the employment-based and family-based preference categories, DV applicants cannot join a long backlog and wait their turn for a visa to become available; instead, applicants must reenter the lottery each year and hope that their number is pulled.

Despite the Trump administration’s rhetoric, winning the lottery alone does not guarantee admission to the U.S.—it only makes one eligible to continue the visa application process. 12 Diversity Visa Program—If You Are Selected, U.S. Dep’t State, https://perma.cc/GJ85-KRCF (archived Sept. 14, 2018). When an immigrant’s number is called, so to speak, they proceed to the nearest U.S. embassy or consulate to submit their supporting documentation. 13 The documents must be brought to the “visa interview at the U.S. Embassy or Consulate.” Diversity Visa Program—Prepare Supporting Documents, U.S. Dep’t State, https://perma.cc/XXF4-QLHP (archived Sept. 14, 2018). The prospective immigrant must then show she has either a high school diploma (or the equivalent), or at least two years of experience working in an occupation that requires at least two years of training. 14 See 8 U.S.C. § 1153(c)(2). But the inquiry does not end there. Next, the review for suspected terrorist ties, criminal record, communicable illnesses, and the like is conducted. Courts grant consular officials significant deference in making these determinations, in a doctrine called “consular nonreviewability.” 15 See Kleindienst v. Mandel, 408 U.S. 753, 770 (1972) (refusing to look behind the executive branch’s decision regarding whether to grant a visa). All this goes to show: The Trump administration’s description of the program does not comport with reality.

II. What Role Does the DV Lottery Play in the Labor Market?

The story of immigration is not just about family reunification; instead, more than half of the immigrant visa categories deal with employment-based or family-based eligibility. 16 See generally 8 U.S.C. § 1101(15) (laying out the categories of immigrant visa, nearly all of which are either based on employment or family relationships, with the exception of the diversity visa and several other smaller programs). For a more accessible format, see Directory of Visa Categories, U.S. Dep’t State, https://perma.cc/488S-HFDH (archived Sept. 14, 2018). In other words, strengthening the labor market is one of the core goals of the immigration system. 17 Jeb Bush et al., Council on Foreign Relations, Independent Task Force Report No. 63, U.S. Immigration Policy 13-20 (2009), https://perma.cc/WTY5-89CQ (describing the broad goals of immigration policy, including, among other things, maintaining the U.S.’s economic dynamism). Many have dismissively suggested that the DV program has no role to play in this regard, 18 See, e.g., Peter H. Schuck, Why the ‘Diversity Lottery’ Needs to End, N.Y. Times (Nov. 2, 2017), https://perma.cc/5J3F-THFD . but this is not the case. 19 See Kelly Virella, Diversity Visa Lottery Winners Defend and Critique Maligned Program, N.Y. Times (Nov. 3, 2017), https://perma.cc/34RH-QBUP (showing that some DV lottery winners have far more than the minimum qualifications for the visa); see also Ryan Radia, A Highly-Skilled Opportunity for Trump on Immigration, Competitive Enterprise Inst. (Jan. 29, 2018), https://perma.cc/A5YC-5KRR (“[M]any immigrants who are admitted into the United States under visa categories that don’t emphasize skills are nevertheless college graduates.”). The DV lottery creates migration flows from areas where the U.S. would otherwise have virtually no immigration. Though not immediately obvious, these flows play a strong role as a “pull factor,” encouraging more settlement, particularly from countries where few migrants with significant skill or education typically emigrate. The DV program thus picks a random sample of trail-blazing immigrants who create the paths future high-skilled immigrants follow.

The DV program creates pathways for future immigrants because it generates a critical mass of people seeking specific amenities and may signal the possibility of job opportunities to potential future migrants. The reasons for seeking a better life in a different country are as varied as the people who choose to make those decisions, but some common themes emerge from this human tapestry of resettlement. First, people tend to resettle in areas where there are other people that come from the same country. 20 See Ann P. Bartel, Where Do the New U.S. Immigrants Live?, 7 J. Lab. Econ. 371, 372, 380-82 (1989) (describing empirically the fact that immigrants tend to follow immigrants from the same country of origin when settling in a new country); Per-Andres Edin et al., Ethnic Enclaves and the Economic Success of Immigrants—Evidence from a Natural Experiment 2-3 (IFAU – Office of Labour Market Policy Evaluation, Working Paper No. 2000:9, 2000) (describing formation of ethnic enclaves via immigration generally, reviewing the literature before applying the observations to Sweden). My line of analysis draws on the theoretical model Enrico Moretti established with respect to city-sorting. Enrico Moretti, Estimating the Social Return to Higher Education: Evidence from Longitudinal and Repeated Cross-sectional Data, 121 J. Econometrics 175, 176-77, 180-184 (2004). Moretti describes inequality in cost of living between cities as a combination of preferences for specific amenities and need for a worker’s skills in a particular location. Id. Although Moretti’s model is not grounded in the context of immigration, the same logic applies to immigration. See David Albouy et al., Immigration and the Pursuit of Amenities 1-3 (Sept. 10, 2018) (unpublished manuscript), https://perma.cc/4RKT-4SRR (citing Enrico Moretti, Real Wage Inequality, 5 Am. Econ. J.: Applied Econ. 65 (2013)) (applying Moretti’s model to immigrant location choices). At a human level, this makes sense, because people tend to feel more comfortable in areas that feel familiar. 21 Jill H. Wilson & Shelly Habecker, The Lure of the Capital City: An Anthro-geographical Analysis of Recent African Immigration to Washington, DC, 14 Population, Space & Place 433, 440, 444 (2008) (“Feeling comfortable or ‘at home’ is important to the people interviewed.”). It is difficult to untangle the relevance of ethnicity as a stand-alone factor from other amenities, but evidence suggests some combination of location-based factors is important and leads migrants to follow previous migrants from their country of origin. See Bartel, supra note 20, at 372, 380-82 (describing the importance of ethnicity as a potential factor in explaining immigrant location choices); Edin et al., supra note 20, at 2. It should come as little surprise, then, that labor economists confirm this intuition. 22 Of course, they do not phrase this observation in such humanistic terms; rather, they typically discuss the immigrant networks in economic terms—but the general point is the same. See David Card, Immigrant Inflows, Native Outflows, and the Local Labor Market Impacts of Higher Immigration, 19 J. Labor Econ. 22, 43 (2001); Bartel, supra note 20, at 372, 380-82; Edin et al., supra note 20, at 2-3. Further, people tend to move to areas that have services and amenities that match their preferences, all else being equal. 23 There is a long stream of research in this particular area, stemming back to Charles Tiebout’s original article describing “voting with your feet” between local communities. See Charles M. Tiebout, A Pure Theory of Local Expenditures, 64 J. Pol. Econ. 416, 420 (1956). If there are virtually no people from a specific country, or even region, in a city, then there is little reason for businesses to spring up offering goods and services catering to the preferences of that specific country or region. 24 See Barry R. Chiswick & Paul W. Miller, Do Enclaves Matter in Immigrant Adjustment? 4-8 (IZA Discussion Paper No. 449, 2002) (explaining that establishing ethnic enclaves produces an economies of scale effect with respect to ethnicity-specific amenities and markets); cf. Enrico Moretti, The New Geography of Jobs 124 (2012) (explaining that the “forces of agglomeration”—the presence of skilled workers with specific training, specialized service providers, and an exchange of ideas—accounts for companies’ decisions regarding where to locate). On the other hand, if a settlement pattern begins to emerge, ethnic enclaves may form, attracting increasing numbers of people from the same country or region, in part because their favorite amenities from home are now available in the U.S. 25 There is even some evidence that enclaves may improve labor market outcomes empirically for some immigrants, although this result tends to be somewhat agnostic as to the causal channel through which this effect operates. See Per-Anders Edin et al., Ethnic Enclaves and the Economic Success of Immigrants—Evidence from a Natural Experiment, 118 Q. J. Econ. 329, 331, 349-51, 354 (2003) (describing the results of a “natural experiment” in Sweden and concluding that ethnic enclaves may have high levels of ethnic income or ethnic self-employment). These enclaves also reduce the uncertainty involved in moving to a new country, as the concentrated population of immigrants may form information networks that help newcomers find employment opportunities. 26 See Chiswick & Miller, supra note 24, at 4 (“Not being connected to host country information networks when they arrive, immigrants have an incentive to create or ‘import’ information networks through living in geographic concentrations with other new and longer term immigrants from the same origin.”); see also Thomas Bauer et al., Herd Effects or Migration Networks? The Location Choice of Mexican Immigrants in the U.S. 3-4, 8-12 (IZA Discussion Paper No. 551, 2002). While subsequent immigrants may not utilize the diversity visa, the DV program plays an important part in enticing future applicants in the merit-based visa categories. The result is that immigrants tend to proceed along well-worn paths, following their countrymen into the same cities and neighborhoods.

A. How Settlement Patterns Take Root: A Case Study of Mexican Settlement in Smyrna, Tennessee

Experience in the Southeastern automobile manufacturing corridor sheds some light on how these settlement patterns begin. 27 See generally David Card & Ethan G. Lewis, The Diffusion of Mexican Immigrants During the 1990s: Explanations and Impacts, in Mexican Immigration to the United States 193, 193-94 (George J. Borjas ed., 2007) (describing “[t]he inflow of Mexican immigrants to Southeastern cities”); Mike Randle, The Southern Auto Corridor: The Past, Present and Future of the Center of North America’s Automotive Universe, Southern Bus. & Dev., Spring 2013, at 14-15, https://perma.cc/R5LP-PTQL (describing the expansion of the Southern Auto Corridor). In 1983, Nissan began manufacturing cars in Smyrna, Tennessee. 28 Randle, supra note 27, at 15. Prior to the plant opening, virtually no Mexican immigrants settled in the area; after the plant opened, immigration exploded, increasing Mexican immigration from near-zero to around 15,000 in just a decade. 29 See Patrick Kennedy, Patterns of Migration Outside Origin Cities: Evidence from the Automobile Industry in the South from 1990-2000, at 10-13 (Aug. 13, 2018) (unpublished Ph.D. dissertation, University of Illinois-Chicago) (on file with author) (describing this in the context of Mexican-born immigrants moving to Smyrna’s metropolitan statistical area). Part of the explanation is obvious: People initially moved to the region to take advantage of jobs in the automotive sector. The story does not end there, however, as many of the immigrants settling in the area did not work in automobile manufacturing; instead, immigrants likely took a number of other positions. 30 This is an inference from data showing that the Mexican population increased by roughly 15,000, although the number of auto manufacturing workers only increased by around 150 in Smyrna’s metropolitan statistical area. See id. at 4, 10-13. A quick infusion of jobs available for immigrants thus creates a magnet for subsequent immigration where none previously existed. 31 See Kennedy, supra note 29, at 4-6, 10-13.

Creating more moments like the Smyrna experience requires that immigrants have the ability to take advantage of emerging opportunities, without major delays while visas are processed. 32 Visa retrogression, meaning longer delays before a visa becomes available for an applicant, may be problems for the employment-based and family-based visas. See Visa Retrogression, U.S. Citizenship & Immigr. Servs. (Mar. 8, 2018), https://perma.cc/7BQB-S7HT (acknowledging the delays in processing employment-based and family-based visas). The DV program requires an annual application, which means the probability of receiving a diversity visa in any given year is fairly stable; in contrast, eligible immigrants in the employment-based and family-based categories are placed in a backlog and often wait many years. See Can I Apply for the Lottery Every Year?, Am. Dream, https://perma.cc/6UL8-ZMMV (applicants must apply each year for the DV lottery, because the government does not roll over applications). This is precisely why the DV program is essential: It creates the possibility of settlement patterns from far-flung corners of the globe, even where none previously existed, breaking the inertia of existing migration patterns. All it takes is one opportunity and a few well-placed folks to get the ball rolling.

To the Trump administration, the Smyrna example might seem like a platform for lamenting (the strawman of) domestic workers losing out on factory jobs in the favor of immigrant labor. But the reality of Smyrna undercuts that argument. First, it appears that not all of the jobs went to immigrants. 33 This is difficult to assess precisely, given the imprecision of the American Community Survey data, but estimates suggest only a portion of manufacturing-specific workers were immigrants. See Kennedy, supra note 29, at 11-12. Second, the availability of immigrant and native labor frequently produces a positive feedback effect, meaning that adding more immigrants does not mean losing American jobs. Immigrants increase productivity 34 See Giovanni Peri, The Effect of Immigration on Productivity: Evidence from U.S. States, 94 Rev. Econ. & Stat. 348, 352-55 (2012) (concluding that immigrants create positive spillover effects in productivity across the skill distribution). and produce roughly net zero effects for American workers as well. 35 See Joseph G. Altonji & David Card, The Effects of Immigration on the Labor Market Outcomes of Less-skilled Natives, in Immigration, Trade and Labor 201, 221 (John M. Abowd & Richard B. Freeman eds., 1991) (describing small negative wage impacts from increased immigration on low-skilled segments of the labor force); Peri, supra note 34, at 352 (“This confirms previous studies . . . that report no evidence of crowding out of native employment by immigrants using correlation across local labor markets.”). Thus, immigrant enclaves such as Smyrna bring dual benefits of immigration pull factors and net increases in productivity.

B. How Settlement Patterns Attract High-Skilled Immigrants: A Case Study of Nigerian and Ethiopian Settlement in Washington, D.C.

The historical move of African immigrants from Europe to the U.S. illustrates the labor market advantages of the DV program. “Diversity visas are one of the few ways people from Africa and the Caribbean can come to this country,” 36 Pamela Constable, Green Card Lottery, A Ticket to Hope for Many, Could Be Eliminated, Wash. Post (May 12, 2013), https://perma.cc/KF5L-TNT5 (quoting Representative Donald Payne Jr.). and they therefore form an important part of the story of creating the conditions for continuous migration from Africa across the skill spectrum. 37 See April Gordon, The New Diaspora-African Immigration to the United States, J. Third World Stud., Spring 1998, at 79, 99 (“[P]rimary migrants arrive first. After becoming residents and citizens, they bring their families over. Those families in turn become residents and citizens . . . . This encourages even more of their countrymen and countrywomen to come, and so the process continues.”).

While “push factors” have led to roughly 30% of Africans with bachelor’s degrees leaving the continent, their primary destination has traditionally been Europe. 38 Id. at 86 (“From 1960 to 1989, an estimated 70,000-100,000, highly skilled African workers and professionals left their countries to go to Europe or, secondarily, the United States. . . . [A] third of all college graduates have left the continent.”). But concurrent with the adoption of the DV program, Washington, D.C. saw an expansion in Ethiopian and Nigerian immigration in the 1980s, which took off after 1990, with many immigrants settling in areas with a majority-black population. 39 Samantha Friedman et al., Race, Immigrants, and Residence: A New Racial Geography of Washington, D.C., 95 Geographical Rev. 210, 217-19 (2005); African-Born Blacks in the Washington, D.C., Metro Area, Population Reference Bureau (Jan. 24, 2008), https://perma.cc/K4KD-RL58. “While the population of Africans in the US is still small compared with most other immigrant regions of origin, it has grown rapidly, especially since 1990 . . . 70% of the 1.25 million foreign-born Africans in the US in 2005 had arrived within the previous 15 years.” Wilson & Habecker, supra note 21, at 437. Subsequently, African immigration grew to more than 11% of D.C. immigration, compared to a national average of less than 3% 40 Friedman, supra note 39, at 211. Current estimates place the number of African immigrants in Washington, D.C. at almost 17% of immigrants, with nearly half of them arriving after the year 2000. See Facts on the District’s African Community, D.C., https://perma.cc/H8XH-6G4H (archived Sept. 24, 2018). Today, the African countries with higher numbers of DV recipients have higher-skilled immigration streams, as opposed to African countries with higher numbers of refugees or more family-based immigration. 41 See Randy Capps et al., Migration Policy Inst., Diverse Streams: Black African Migration to the United States 1 (2012). The DV program has been responsible for the growth in highly-skilled immigration from Africa, particularly those in professional and technical occupations. See Ann M. Simons, African Immigrants Are More Educated Than Most—Including People Born in U.S., L.A. Times (Jan. 12, 2018, 7:15 PM), https://perma.cc/8DB7-WMLL (“[T]he influx [from the DV lottery] includes many immigrants from sub-Saharan Africa who are highly skilled professionals.”). This demonstrates the promise of the DV program: Where previously high-skilled immigrants settled in Europe, following behind their countrymen, “[o]nce migrant paths are established, the movement of other migrants becomes easier because there are networks of kin and friends . . . . Services for migrants also become established. All of these factors combine to encourage an acceleration of migration.” 42 Gordon, supra note 37, at 82. The results are clear in D.C. enclaves today, where Ethiopian immigrants are attracted to “the region’s large Ethiopian community who have settled here through the visa program.” 43 See Constable, supra note 36; see generally Bartel, supra note 20, at 380-381. The end result is greater high-skilled immigration from African countries, such as Nigeria and Ethiopia. 44 See U.S. Diversity Visas Are Attracting Africa’s Best and Brightest, Population Reference Bureau (July 1, 2001), https://perma.cc/2HFQ-4XMH (“The diversity visa program . . . provided a swift path of entry to African [immigrants in professional, managerial, and technical occupations] who could not qualify for a visa through . . . family ties. . . . Thus, the diversity visa program became the primary vehicle for the increased outflow of skilled Africans to the United States.”).

The takeaway from the Ethiopian and Nigerian enclaves in Washington, D.C. is this: In order to attract the best and the brightest from their home countries, it helps to have a bedrock of immigration from enterprising people with less to lose from moving their lives overseas.

III. Why Immigrants Cannot Just “Stand in Line”

It is tempting to assume that immigrants from high-immigration countries can simply “fill in” for any loss in immigration—and that there is thus no need for the DV program—but this ignores the strict limitations the Immigrant and Nationality Act imposes on legal migration, particularly employment-based immigration. First, immigrants must meet stringent qualifications to receive lawful permanent residence, such as demonstrating international excellence in their occupation, or possession of an bachelor’s or advanced degree, or years of experience. 45 See 8 U.S.C. § 1153(b) (2016). The EB-3 subcategory includes a cap of 10,000 visas for unskilled workers, which is the lowest preference category. Id. § 1153(b)(3)(B). Second, even when immigrants meet all statutory qualifications, there must be a visa “available,” which is unlikely in many cases, given extreme backlogs and low caps for the most commonly used visas, such as the H-1B, EB-1, EB-2, and EB-3 visas. 46 See Visa Bulletin for June 2018, U.S. Dep’t. State (May 7, 2018), https://perma.cc/PDQ6-MUHB (showing visa retrogression dates for employment-based visas for immigrants from certain countries); Stuart Anderson, Will Congress Ever Solve the Long Wait for Green Cards?, Forbes (May 21, 2018, 12:05 AM), https://perma.cc/C5U8-J7SX ; The Latest Visa Bulletin, Explained, Boundless (Sept. 13, 2018), https://perma.cc/6G3M-QBWG. The per-country cap further exacerbates the shortfall, since the cap limits the number of available family-based and employment-based visas for each country. 47 See 8 U.S.C. § 1152(a)(2) (limiting “the total number of immigrant visas made available to natives of any single foreign state or dependent area” to 7% each fiscal year). In other words, there could be years when the overall visa cap is not hit, but when too many high-admission countries hit their per-country quota.

Even if Congress removed the per-country cap, which it certainly will not do, eventually all of the best matches from high-admission countries could apply and be admitted, 48 Of course, it is also possible that the growth in the population interested in migration would outstrip the U.S. immigration system’s ability to process visas, even in a world with neither a per-country or per-category cap on immigration. Estimating empirically which of these competing theories is correct is difficult. Regardless, the point remains that it is politically infeasible to remove either cap, both of which have existed since the early days of the Immigration and Nationality Act, passed in 1965. leaving fewer interested and qualified prospective additions to the labor market. 49 This is the result of two trends in immigration from India and China. First, immigrants in lower-end qualifying positions in high-technology industries, such as computer programmers, face additional application barriers and are choosing to go to Europe, Canada, or other more hospitable places. See Ananya Bhattacharya, Indian Techies Look to Canada as the American Dream Turns into an H-1B Nightmare, Quartz (Feb. 26, 2018), https://perma.cc/7HCN-D4W4 (discussing Indian immigrants already in the U.S. leaving for Canada).; see also Julia Funke, Supply and Demand: Immigration of the Highly Skilled and Educated in the Post-9/11 Market, 48 J. Marshall L. Rev. 419, 437 (2015) (“[T]his advantage is quickly disappearing as other countries relax their immigration systems to make it easier for highly skilled workers to come to their countries at the same time the U.S. is making its system more restrictive than ever,” causing “many foreign-born workers to settle elsewhere.”). Second, as economic conditions improve in China and India, the wage differential between working in the U.S. and remaining in China or India may shrink. Should this occur, immigrants are unlikely to bear the increasing costs and uncertainty of applying to immigrate to the U.S. when they face diminishing returns in terms of wage gains. See Chiswick, supra note 6, at 2-3. In other words, even if everyone that wanted to emigrate from China and India were exempt from all caps and quotas, eventually all interested and qualified candidates would move to the U.S., particularly as economic conditions improve in their countries of birth. Put differently, not everyone wants to leave their home country; in fact, most people want to stay. This means we cannot expect to keep attracting talent from the same pool of countries indefinitely. Eventually, those pools will run dry. 50 In fact, with respect to workers from the U.S. and many other developed nations, high-skilled labor availability is declining. See Karen Harris & Andrew Schwedel, Automation, Older Workers to Shape US Workforce in 2020s, The Hill (May 17, 2018, 11:30 AM EDT), https://perma.cc/9XR5-BLQY; Beige Book – January 17, 2018, Board Governors Fed. Res. (Jan. 17, 2018, 2:00 PM EST), https://perma.cc/9XKA-8C3U (“Most Districts cited on-going labor market tightness and challenges finding qualified workers across skills and sectors, which, in some instances, was described as constraining growth.”); see also Hays plc, Regional Dynamics of the Global Labour Market: Skills in Demand and Tomorrow’s Workforce 4 (2017), https://perma.cc/8WE5-QCPQ (noting that organizations across the globe “continue to deal with skills shortages”).

Even when significant factors push immigrants to the U.S., such as a long history of migration, unrest, or civil strife, the DV lottery still plays an important role. While in these cases the combination of push and pull factors likely would propel immigrants, neither resolves the issues identified here. First, the per-country caps work strongly against Indian, Chinese, and Pilipino immigrants, which have long histories of sending emigrants to the U.S., forcing them to wait years for visas. 51 See The Latest Visa Bulletin, Explained, supra note 46 (providing updates on the general backlog for employment-based and family-based visas); Anderson, supra note 46 (explaining that some Indian immigrants can face wait times of between ten and twenty-five years). Second, countries experiencing distress typically produce significant numbers of refugees who either do not meet requirements for such visas or cannot afford to wait. 52 Some refugees cannot wait because they are fleeing persecution. See Kalena E. Cortes, Are Refugees Different from Economic Immigrants? Some Empirical Evidence on the Heterogeneity of Immigrant Groups In the United States, 86 Rev. Econ. & Statistics 465, 465, 471 (2004). Though these refugees may be eligible for asylum and able to stay in the country, in many cases they will not be eligible for asylum (for any number of reasons, including strict procedural bars) and will not qualify for entry. 53 See Linda Camp Keith & Jennifer S. Holmes, A Rare Examination of Typically Unobservable Factors in US Asylum Decisions, 22 J. Refugee Stud. 224, 225-28, 229 & tbl.1, 230 (2009) (describing increasing restrictions in the asylum process and showing that the national rate of denial is around 60%). Even if they do, refugees may not be self-selected the same way as economic migrants, because they are fleeing persecution, rather than affirmatively seeking economic opportunities. 54 See Cortes, supra note 52, at 465.

In other words, although it seems like “there are always people willing to fill in,” the reality is that those streams will eventually run dry without new sources of labor running into them. The DV program is the obvious solution: Lottery spots create flows of immigrants from new sources; those immigrants then begin to create ethnic goods and services, including information networks, attractive enough to help pull the highly skilled into the U.S. Thus, the DV system creates the necessary preconditions for the U.S. to serve as a global magnet for the best and brightest—because without it, as history shows us, highly-skilled individuals remain at home.

The DV program plays an important role in the long-term strategy for maintaining the competitiveness of the U.S. labor force. Without additional streams of immigrants looking to resettle, per-country caps on immigration and an eventual lack of interested, qualified candidates from high-admission countries would set the U.S. back competitively. 55 See Funke, supra note 49, at 435-37 (arguing that the visa caps are responsible for much of the decline in skilled immigration, which in turn harms U.S. competitiveness); see also supra note 48 (explaining caveats to the theoretical argument that there could eventually be no interested, qualified candidates). Even though many qualified candidates for admission come from the two most populous countries, India and China, both nations also suffer from major backlogs caused by the per-country cap and visa caps. 56 See Carla N. Argueta, Cong. Research Serv., Numerical Limits on Permanent Employment-Based Immigration: Analysis of the Per-country Ceiling 7 (2016) https://perma.cc/Y7WE-BR9G (“Current” in this source means a visa is presently available, while “priority date” means the date at which one would have needed to apply to receive a visa, based on the most recently approved application.). This is also true for Mexican immigrants with respect to family-based visas. See Samuel W. Bettwy, A Proposed Legislative Scheme to Solve the Mexican Immigration Problem, 2 San Diego Int’l L.J. 93, 97 (2001). These caps prevent quick and mobile movement of many workers into the U.S. labor market, necessitating further sources of labor. 57 See David Bier, 150-Year Wait for Indian Immigrants with Advanced Degrees, Cato Inst. (June 8, 2018, 12:45 PM), https://perma.cc/5Q4W-5Q6N (wait times can stretch on for decades for Indian immigrants because of per-country caps); cf. Anderson, supra note 46 (explaining reduced mobility in the context of immigrants who cannot leave their original employers without first receiving a green card).

Conclusion

Not all of the proposals recently before Congress would cut the DV program: Several bipartisan immigration reform efforts would maintain visa eligibility at current levels and also keep the DV program. 58 Desjardins, supra note 2. The Republican reform plan included a modified version of Representative Goodlatte’s proposal, which would end the DV program and restricted family migration. 59 Don Wolfensberger, House GOP Leaders Fail to Find Compromise Immigration Fix, The Hill, (July 3, 2018, 7:30 AM EDT), https://perma.cc/5T78-ELHL. Even this plan faced significant political difficulties and the possibility of a presidential veto. 60 It remains unclear whether President Trump will make good on his promise, since he has waffled on his threat to veto. Scott Detrow, Latest Republican Immigration Bill Fails to Win Approval in the House, NPR (June 27, 2018, 5:27 PM ET), https://perma.cc/7VJU-S3NV (“But within the last two weeks alone, the president said he might veto this bill. Then, he said he supported it. Then, he said it was a waste of time.”). Given recent statements that President Trump is willing to play hardball with Democrats over immigration, a plan that does not end the DV program may have little chance of becoming law. 61 See Erin Corbett, Trump Threatens to Shut Down Government If Democrats Don’t Back Immigration Policies, Fortune (July 29, 2018), https://perma.cc/3GZ5-JLH5. The Democrats and moderate Republicans should stand strong against such proposals in light of the strong labor economics case for maintaining the DV program. They should not trade away the future immigration networks trailblazers from low-admissions countries will build through the DV lottery.

J.D. Candidate, Stanford Law School, 2019. PhD Candidate, University of Illinois-Chicago Department of Economics, 2019. Thanks to my family for reading and suggesting changes to the piece, especially Denise DeArmond, Sean, Chris, Michael, and Joan Kennedy. Thanks to the Stanford Law Review Online staff for their helpful comments and guidance, especially David Hoyt and Andra Lim.