Review of The Culture Transplant: How Migrants Make the Economies They Move to A Lot Like the Ones They Left by Garett Jones, Part I
The First Installment
This is the first part of a long review and essay on The Culture Transplant: How Migrants Make the Economies They Move to A Lot Like the Ones They Left by Garett Jones, Associate Professor of Economics at George Mason University. Jones’s book has earned several interesting reviews, endorsements, and criticisms since being released in November.
Helen Andrews at the American Conservative asks, “Is Garrett Jones kidding? He has written a book so pro-immigration in its argument, yet so anti-immigration in its substance, that I cannot discount the possibility that he is engaged in a sophisticated satire.” Bryan Caplan argues that Jones's book was just a few edits away from being a pro-immigration book that would support "50 percent open borders." George Francis praises Jones for “fairly consider[ing] all criticism,” but criticizes him for being light on theory. Jason Richwine focuses on how low trust among immigrants “devastates the starry-eyed ‘come one, come all’ immigration advocacy popular among progressives and libertarians.” Finally, Jones’s colleague Tyler Cowen blurbs the book as “the very best book on this phenomenon.”
Except for Cowen’s praise, my reaction overlaps with the above reviews and adds another point: The Culture Transplant is incomplete. Jones left so much unwritten, so much research unmentioned, and so many counterarguments unexamined that the book reads as if it were incomplete.
These reviews (essays?) will examine the academic peer-reviewed literature that Jones excluded from his book, consider other conclusions one can draw from his presentation of the literature, and present some additional points he left unexamined. A references section at the end of this review will provide further reading for those interested.
Background
Social scientists have responded to the increased political saliency of immigration by producing reams of interesting, unique, and valuable social science research on this topic. The standard economic finding is that immigration restrictions impose many trillions of dollars of deadweight loss annually (Clemens, 2011). A subtopic in this vast area focuses on determining how immigrants will affect the productive capacity of the countries where they settle. This subtopic is vital because, depending on the findings, the deadweight loss from immigration restrictions could increase, shrink, or even become negative depending on how immigrants affect the productive potential of destination countries.
I have a deep interest in this question because it is the best potential counterargument against free immigration. As a result, I’ve co-authored a few peer-reviewed papers and one book on how immigrants affect economic and political institutions in destination countries — basically starting this literature with a simple blog post that hasn’t migrated well.
How immigrants impact the productive capacity of their new homes is now the most plausible anti-immigration argument because the case for massively expanded legal immigration is so robust. The effects of immigrants on labor markets, innovation, entrepreneurship, crime, terrorism, government budgets, language, and myriad other issues are well studied. Of course, immigrants impose some costs, higher in some places and lower in others, but the overall impact is hugely positive, especially when including the welfare of the immigrants and their descendants. As a result, smart skeptics of immigration have moved to other criticisms: How immigrants will affect the fundamental cultural, institutional, and political factors that determine a society's productive capacity over the very long term.
Garett Jones is one such smart skeptic. He argues that immigrants and their descendants don’t assimilate fully into the cultural norms of their destination countries even after many generations. Jones focuses on cultural assimilation measures like generalized trust and how ancient history affects prosperity today. Jones concludes that immigrants and their descendants will transform the economies of destination countries into being much like the poor countries where they came from. The long-term result, he claims, is that immigrants would kill the goose that lays the golden eggs of economic growth by eventually degrading the productive and innovative capacity of destination countries. As a result, in the long run, the world would be substantially poorer with freer immigration than with current restrictions.
Jones’s book is ambitious. He’s standing athwart much of the peer-reviewed academic literature on immigration and saying, “You’re wrong.” Jones performs a valuable task here because academic literature can often be mistaken, and errors can persist for decades or longer. Jones repeatedly writes that he “kicks the tires” of the papers he cites to guarantee that they are sound.
On the other hand, Jones’s book is incomplete. It largely ignores peer-reviewed research that contradicts his thesis. What Jones claims is “tire kicking” is often just robustness checks run by the authors of Jones's preferred papers. Worse, Jones didn’t catch several large nails stuck deep in the rubber of at least one of his favorite papers. Jones’s unfamiliarity with the vast peer-reviewed literature is not one-sided – it is not purely in service of his thesis. Perplexingly, Jones also ignores peer-reviewed research that would have bolstered his thesis. This is evidence of his incomplete reading of the relevant research, it is not evidence of “p-hacking” his literature survey. As a result of bringing so much research together without thoroughly reviewing its critics, Jones builds parallel intellectual rabbit holes that occasionally cave in on each other.
Jones also doesn't demonstrate that immigrants transform their new countries to be like their old ones. He could have remedied this oversight by including original empirical work instead of focusing on distilling a lesson from an incomplete literature survey. Although one reviewer did take Jones to task for failure to identify a theory or mechanism when Jones’s earlier book Hive Mind provides one, this criticism is slightly unfair. Jones did consider a few mechanisms, mainly trust, but he could have evaluated them better.
Most readers will understand Jones's book to be anti-immigration. However, he could have used the same evidence to support massively liberalized immigration, about 50 percent free immigration according to Bryan Caplan's review. Jones also doesn’t consider a colossal expansion of guest worker visas that would allow migrants to work temporarily in developed countries before returning home. Singapore, the United States, the Gulf Countries, and others have or have had large guest worker visa programs. These programs are compromises that capture many economic gains from immigration while avoiding the potential long-term costs that worry Jones.
Omitting a discussion of guest workers is odd because Singapore has a huge guest worker population and The Culture Transplant is the third book in Jones’s self-titled “Singapore Trilogy.” It’s a bad sign that a trilogy that extols the benefits of Singaporean governance and culture ends with a book about immigration that doesn’t consider Singaporean immigration policy. If the United States had a guest worker visa program on Singapore’s scale, there’d be about 100 million guest workers here — about 50 times the current number.
Two Meta Weaknesses
The chapters in Jones’s book are each built around a handful of interesting peer-reviewed papers about the persistence of immigrant cultural attitudes. There are two overall major weaknesses. The first is a lack of original empirical or theoretical work to plug the gaps in the literature. This is especially egregious because several of the papers he relies on are not about immigration specifically. The second major weakness of Jones’s book is that he ignores much of the vast empirical economics literature.
The second weakness contrasts with Jones’s earlier books Hive Mind and 10% Less Democracy, which seem to carefully evaluate the academic literature. His two earlier books suffer slightly by failing to fill empirical gaps in the literature, but there were probably fewer gaps to fill. My expectations were thus high that Jones would similarly parse the complex academic literature on how immigrants affect host countries' cultures.
Jones on Trust: Can You Trust Him?
Jones identifies generalized social trust (trust) as one of the few cultural mechanisms that can explain how immigrants degrade economic growth in destination countries. He presents a 2010 paper by Yann Algan and Pierre Cahuc (A&C) about the persistence of immigrant cultural attitudes over generations in the United States. A&C (2010) use responses from the World Values Survey (WVS) and the U.S. General Social Survey (GSS) over generations to estimate how trusting young immigrants were when they arrived a century ago. They then purport to show that the descendants of immigrants have not moved much toward the American norm of higher trust compared to their ancestors’ countries of origin in 2000. A&C (2010) then link that lack of assimilation to lower economic growth in 2000.
To further understand A&C (2010) and Jones’s presentation of trust as a mechanism for affecting growth, it's important to grasp what economists mean when they write about “trust.” Economists have been long interested in how culture affects economic growth, but culture is a big topic. What aspects of culture are economically meaningful? Which are quantifiable? As a result, economists face the challenge of incorporating culture into economic models and often treat it as a black box where inputs enter the box and economic outputs exit after being transformed in the dark (Torsvik, 2000). Many social scientists spun their wheels on social capital theory to analyze culture's economic effects. However, the social capital literature is a mess of “unclear and incoherent definitions . . . that is even vaguer when it comes to specifying the mechanisms that link social capital and production . . . it is fair to say that even though social capital might be the missing link in economic development, what is more apparently missing is a clear specification of the links between social capital and production” (Torsvik, 2000, p. 452). To fill the cultural black box, some economists concentrate on measures of trust as a proxy for economically relevant culture (Gambetta, 2000).
Trust is “the subjective probability with which an agent assesses that another agent or group of agents will perform a particular action” (Gambetta, 2000). Economists have settled on trust as an economically relevant cultural trait for two main reasons. The first is that trust seems like it can be incorporated into standard economic models, although it rarely is (Algan & Cahuc, 2010, 2013; Guiso, Sapienza, & Zingales, 2009b; Zak & Knack, 2001). The second is data availability (Weil, 2005). Surveys like the WVS, EuroBarometer, GSS, Latinobarómetro, and others have asked similar questions about trust for decades in different countries.
The specific trust question used in the WVS asks respondents: “Generally speaking, would you say that most people can be trusted or that you need to be very careful in dealing with people?” The responses are “Most people can be trusted,” “Can’t be too careful,” and “Depends.” Economists interpret the response “Can’t be too careful” as the respondent possessing low trust and “Most people can be trusted” as the respondent having high trust. The GSS question is nearly identical.
Jones claims “that economists have devoted vast theoretical and empirical efforts to show how trust and its more important twin, trustworthiness, shape the wealth of nations.” He’s correct that there is a vast empirical literature, but there are few theoretical microeconomic models of how trust affects production on the firm level and none for modeling the effect of trust on the macroeconomy. The trust literature is not theoretically rigorous, and that's one of its weaknesses. However, there are plenty of informal models based on stories and some micro models that I’ll dive into below.
The big problem is that Jones presents A&C (2010) as settled social science while ignoring its critics. The first such criticism was by Müller, Torgler, and Uslaner (2012). A&C (2010) calculated levels of trust early in the 20th century by backcasting trust based on immigrant responses in earlier survey waves from the 1970s. Müller et al. (2012) found that the sample sizes of immigrants who arrived in the early 20th century, which is the basis for inherited trust, are small. Specifically, A&C (2010) calculate inherited trust in 1935-38 for those born in 1910. For that calculation, there are 11 respondents of Hungarian descent in the pooled 1972-2004 GSS, 35 of Swedish background, between 40-100 respondents for people of Dutch, African, and Norwegian ancestry, and more than 100 for those of English, German, Irish, and Scottish ancestry. In the words of Müller et al. (2012), “[t]hese are tiny samples in which to base bold conclusions.”
The most-cited regressions in A&C (2010) rely on the 2000 wave of the WVS. The WVS survey is a massive global survey of attitudes in different countries collected in 5-year waves (1999-2004, for instance). The major problem is that the 1999-2004 wave of the WVS produces strange results that don't match well with country surveys like the GSS and EuroBarometer, which have more respondents. In most cases, re-running the analysis by A&C (2010) for WVS survey years other than 1999-2004 reduces the coefficient, reduces or eliminates the significance, and sometimes both. WVS Wave 4 for the years 1999-2004 has significantly different trust responses relative to many other survey datasets to such an extent that the results presented in A&C (2010) were not robust in other waves of the WVS (Clemens & Pritchett, 2019). This lack of replicability for other waves suggests that A&C’s (2010) results may merely be an artifact of WVS mismeasurement in a single wave. Another published criticism by Eder (2018) makes a few adjustments to A&C’s (2010) regression models to improve causal inference and reduce endogeneity and omitted variable biases. Both sets of adjustments produce statistically insignificant results or economically insignificant effects of trust on GDP per capita.
Müller et al. (2012) and Eder (2018) found problems with A&C (2010) that should reduce our confidence in the latter paper, but you wouldn’t know it from reading The Culture Transplant. It’s too bad because, as Jones acknowledges, “[Smart], careful critics can help us test our theories. The adversarial approach to finding truth—where each faction tries to critique and debunk the claims of the other factions—works well in the courtroom and it works well in economics." Unfortunately, it only works well if economists pay attention to the critics.
Beyond that, there are at least five serious problems with the trust literature that Jones should have mentioned. The first problem, as previewed above, is that the trust literature contains no formal economic growth models that incorporate trust (Beugelsdijk & Maseland, 2011, p. 213). As a result, it’s difficult to interpret the empirical results or regressions without a model.
I hypothesize that there are two reasons why economists haven't incorporated trust into growth models. The first is that the informal and micro models incorporating trust are concerned with transaction costs. Incorporating transaction costs into macro growth models would be a tiny parameter adjustment that would ~zero impact. That may be a problem for the growth models, but proponents of the trust-growth hypothesis should address it. My second hypothesis is that economists have attempted to incorporate trust into growth models, perhaps even by editing those models to make trust more important, but the results are still null or trivial, so those results remain unpublished due to publication or other biases.
However, plenty of informal and semi-structural models may be able to identify trust as a mechanism (Blundell, 2017). One informal model assumes that transaction costs are higher when trust is lower, which diminishes growth. Another informal model is that high trust reduces the resources that firms and individuals rationally spend on protective purposes, freeing those resources up for productive uses (Bjørnskov, 2018). A more formal innovative investment model includes social norms, social trust, and networks with reciprocity to model investment in productive ideas that have economy-wide effects (Ikeda, 2008; Akcomak & Weel, 2009). A semi-structural model by Bjørnskov (2009) incorporates trust into firm search costs for educated workers. He identifies trust as a mechanism that increases firm-level demand for educated workers that increases schooling, thus potentially increasing human capital, which matters in some endogenous growth models (Romer, 1986; Lucas, 1988). Butler et al. (2016) is a good paper that recognizes the empirical and theoretical limits of the trust-growth literature. They construct a stylized microeconomic model to explain individual investments in trust and their effect on income, finding an n-shaped relationship where too little individual trust and too much trust are correlated with lower individual income in Europe (Butler et al., 2016). They don’t pretend that their microeconomic model has macroeconomic implications and, thus, don’t claim it has a meaningful impact on national growth or levels of GDP per capita. Another formal micro model embeds a principal-agent situation in a model of education and monitoring costs (Bjørnskov, 2018). Makrychoriti et al. (2021) avoided the theoretical modeling phase when examining the association between financial stress and economic growth. Zak and Knack (2001) come the closest with a stylized investment-driven microeconomic model that includes trust.
Economists have not incorporated these formal models into economic growth models (Rose, 2011, p. 171). Some of the above papers and others rely on Barro-style reduced form regressions that assume an underlying endogenous growth model when they use macroeconomic data to empirically test their microeconomic models. However, Barro-style reduced-form regressions are a poor substitute for a structural endogenous growth model. Those reduced-form regressions preclude analyzing counterfactuals such as immigration-induced effects on trust (Blundell, 2017).
Zak and Knack (2001) provide the most promising path forward for economists who study trust. There could be a solid way to include trust-affected-investment in growth models, but Zak and Knack (2001) stick to their micro model that they test with cross-country macroeconomic data in reduced form regressions. But they face two challenges. One, their theoretical model can only suggest the direction of their empirical results and, at most may be able to identify a transmission mechanism. Maybe. But this cannot be the basis for causal claims about trust's impact on the macroeconomy. Two, reduced-form regressions are very unlikely to incorporate all possible causal growth determinants. They’d need a growth model for that. It says a lot that nobody has incorporated the insights of Zak and Knack (2001) into a model to test the growth impact of trust. Thus, the “empirical literature has proceeded . . . without much clear interaction with theoretical development” (Bjørnskov, 2018). Jones should have kicked this tire.
The second problem is that the trust question does not produce internally valid responses. The meaning of trust is contextually and culturally specific, so the meaning of the responses is unclear (Moore, 1999; Putnam, 2000; Miller & Mitamura, 2003; Hardin, 2006; Beugelsdijk & Maseland, 2011, pp. 216-217). Glaeser, Laibson, Scheinkman, and Soutter (2000) argue that the trust question measures respondent opinions about the trustworthiness of others, not their personal level of trust and that one’s aversion to betrayal is not correlated with trust in cross-country analyses (Bohnet, Greig, Herrmann, & Zeckhauser, 2008). Additionally, responses to the trust question are also not internally valid over time within countries.
The third problem with the trust literature is that responses to the trust question generally don't predict trusting behavior in real-world micro-level experiments, trust games, or investment games (Algan & Cahuc, 2013; Ermisch, Gambetta, Laurie, Siedler, & Uhrig, 2009; Gächter, Herrmann, & Thöni, 2004; Glaeser et al., 2000; Glaeser, Liabson, & Sacerdote, 2002; Karlan, 2005; Näf & Schupp, 2009; Rose, 2011; Sapienza, Toldra-Simats, & Zingales, 2013), with some exceptions (Glaeser et al., 2000; Fehr, Fischbach, Rosenbladt, Schupp, & Wagener, 2002; Felten, 2000). People tend to be more trusting and trustworthy in experiments and games than they would appear to be in surveys (Rose, 2019; Beugelsdijk & Maseland, 2011; Gächter et al., 2004; Holm & Danielson, 2005; Glaeser et al., 2000). Even when these experimental results show that responses to the trust question are correlated (0.34 correlation coefficient) with trustworthy behavior (Glaeser et al., 2000), we should have less trust in cross-country responses to the trust question. Jones's first book, Hive Mind, was thick with interesting psychology and economics experiments, including those about trust. However, the only mention of an experiment in The Culture Transplant concerns the discovery of aspirin in an extended metaphor. This is a glaring omission for those familiar with the trust literature.
The underlying non-formalized story for why trust matters is that countries “characterized by trust are economically more successful is mostly based on micro-level arguments deduced from either transaction costs theory or game theory . . . [and] reduced transactions costs and principal-agent problems . . . increase (the efficiency of) investments in physical and human capital, and promote innovation” (Beugelsdijk & Maseland, 2011, p. 208). The divergence between actions in the laboratory and micro-level experiments and responses to trust survey questions raises significant problems with that underlying story. The surveys could be flawed, the experiments could be flawed, or both could be flawed, but the different results show that at least one of those is true and that calls into question the quality of the trust data.
The fourth major problem with the trust literature is that many of the major papers are contaminated by country sample selection bias, whereby results change significantly based on the sample of countries chosen (Durlauf, 2002). Sample selection bias also occurs when researchers compare the trust level of immigrants in destination countries with the trust level of their former co-nationals in their homelands (Algan & Cahuc, 2010). Emigrants likely have different levels of trust than non-emigrant citizens from their home countries due to self-selection because immigrants are not a random sample of a country's population (Uslaner, 2008a). For instance, the “can trust” response from black African immigrants in the United States is 35 percent, just below the 38.4 percent of white Americans and more than twice that of African American descendants of slavery according to the 2010-2022 pooled GSS sample (warning, small sample size). The WVS survey response “can trust” in the black African countries in the WVS is 9 percent. The difference in “can trust” between black African immigrants and black African non-immigrants is huge. There are caveats, of course, such as the WVS surveys few sub-Saharan African countries, African immigrants come from many countries not surveyed, and the U.S. GSS sample of African immigrants is tiny – but the difference perhaps indicates that self-selection and policy-selection are important. Additionally, emigrant self-selection likely affects measurements of trust in later generations and is even further disturbed by ethnic attrition or the selective and biased self-identification of the descendants of immigrants with ethnic and racial groups (Duncan & Trejo, 2016; Fernandez, 2011; Giavazzi, Petkov & Schiantarelli, 2019; Weil, 2005). Another source of sample selection bias is that many of the findings in this literature depend on the years chosen for study, as Müller et al. (2012) noted.
The fifth problem with the trust literature is that even if the trust question were free from measurement error or sample selection bias, trust might be a proxy measurement for other deeper causes of economic development (Clemens & Pritchett, 2019). Omitted variable bias and endogeneity are persistent unresolved problems in this literature. Robert Putnam said the “causal arrows” in the trust literature are “as tangled as well-tossed spaghetti” (2000, p. 137). Papers in the trust literature often estimate cross-country Barro-style growth regressions using cross-sectional datasets or panel data models (Blume, Brock, Durlauf, & Ioannides, 2010; Knack & Keefer, 1997; Roth, 2009). A critical consideration is whether the parameters underlying these econometric models are truly identified under a set of reasonable assumptions (Blume et al., 2010; Durlauf, 2002). The trust literature heavily depends on instrumental variables (hold your laughter), but none of the major instruments used in the literature pass tests of over-identification and weak instrumentation (Durlauf, 2002; Guiso et al., 2011; Durlauf, Johnson, & Temple, 2005).
A thorough “tire kicking” would have acknowledged and dealt with these problems.
Those five criticisms of the trust literature are listed and described in greater detail in a working paper I coauthored with my former colleague Andrew Forrester (Nowrasteh & Forrester, 2020). Jones dismisses that working paper by writing, “his critiques have yet to find resonance in academic circles.” Fair enough, it’s in R&R purgatory, but 5/6 of our working paper reviews the academic trust literature. We didn’t invent the criticisms. We merely read the literature, came away unimpressed, doubted the claimed link between trust and growth, and decided to combine those peer-reviewed critiques in one place. Our working paper is similar to what Jones did with his earlier books Hive Mind and 10% Less Democracy, but that he failed to do in The Culture Transplant.
The last 1/6 of our working paper gathers the best empirical methods from the trust literature to run a regional analysis of the relationship between trust and income growth in the United States. A regional analysis of the United States removes some, though not all, of the empirical and methodological concerns because trust is regionally heterogeneous under relatively stable and consistent institutions inside of the United States. We did not attempt to add trust into a growth model – proponents of the trust-growth theory should do that. We undertook this analysis after being inspired by Dincer and Uslaner (2010), Tabellini (2008, 2010), and others. We found no relationship between trust and income, even when we used alternative trust measures as independent variables. None of our findings were statistically significant, the signs switched based on the measure of trust, and the magnitudes were tiny. Trust in the United States is declining, as measured by the GSS, and that decline over the 1972-2016 period doesn't correlate with changes in income.
The trust-growth literature is a lot more divided and heterogeneous than Jones admits. Perhaps trust is a sizeable contributor to growth, but the literature doesn’t show that. Anecdotally, it's easy to see why survey measures of trust don't comport with reality. Trust in the United States has plummeted since the 1970s, but it is not declining in the United States on a personal behavioral level – just the opposite. Americans frequently get in the unmarked cars of strangers for rides because an electronic app run by Big Tech says it's ok. That's an industry where low trust should have held back economic activity – “get in a car with a stranger, are you crazy?” But it didn't, despite the admittedly anecdotal experience of most of us being driven around by foreign-born Uber and Lyft drivers. The ubiquity of anonymous online purchases, widely regarded as suspicious in recent memory, is another sign of increasing trust. One could respond that the apps created by the ride-share companies or online payment security increased trust by aligning incentives, but that is still a strike against the data on trust decline changing relevant economic behavior.
Adam of Bremen, an 11th-century chronicler in Northern Europe, commented that the Vikings "have no faith in one another, and as soon as one of them catches another, he mercilessly sells him into slavery." As the Scandinavian economy changed and economic opportunities expanded beyond plunder and slave trading, Scandinavian levels of trust likely changed in response to changing incentives – and now they are among the most trusting people in the world (if you take measures of trust seriously). Sure, we don't have surveys of generalized trust from 1000 years ago, and maybe Adam of Bremen was an anti-Viking bigot, but if incentives changed the trusting behavior of the ferocious Vikings, American Uber drivers and riders, and internet purchasers, then low trust is a solvable problem if it is even a problem. It is certainly cheaper to reduce the hypothesized harm from generalized social trust than to suffer the trillions of dollars of annual economic loss from immigration restrictions.
Jones’s omission of incentives in explaining part of human behavior regarding trust is an oversight, especially noticeable because Jones is an economist. Implicit in much of the trust-growth literature is that trust is a pro-social motivation that some people possess (Uslaner, 2002). But some trusting behavior can also be explained by monetary and reputational incentives, such as players cooperating more in repeated iterations of the prisoner’s dilemma (Torsvik, 2000; Hardin, 2006, p. 68). Those incentives work best when there are clear high costs to opportunistic behavior, such as being fired by Uber or losing online customers. That also means incentivized trusting behavior is more fragile than a pro-social or innate sense of trust, but better incentives can fill at least some of the gap left by a lack of social-solidarity-induced innate feelings of trust. Incentives created by specific institutions and organizations fill that gap in the United States.
Whither Trust Research?
The trust-growth literature is mostly an example of hokum born of desperation. Economists are desperate to explain economic growth, and trust was a simple way to establish a correlation between GDP per capita and a seemingly measurable cultural variable. The theoretical and empirical problems with the trust literature and the well-known criticisms of papers that Jones relies upon should have been part of his “tire kicking.”
But are the methodological, theoretical, and data problems severe enough that all trust-related research should be jettisoned? I don't think so, but trust researchers must recalibrate their research to comport with theory and better empirical evidence. One place to look is trust in specific institutions and organizations. Chapter 8 of my book Wretched Refuse? The Political Economy of Immigration and Institutions (coauthored with Benjamin Powell) discusses differences in trust between immigrants and natives.
We use pooled General Social Survey (GSS) data of responses to numerous trust questions for the 2008-2018 period. Thirty-three percent of native-born Americans say they “can trust” in the classic trust question, compared to 23 percent of immigrants – a large and statistically significant difference. Similarly, 63 percent of natives and 71 percent of immigrants say they can't trust – another large and statistically significant difference (Nowrasteh & Powell, 2021, p. 174). Interestingly, this relationship flips when you ask about trust in specific institutions. We’re not the first to undertake that research, but there is much less of it than there should be (Nannestad, 2008, p. 423).
Immigrants have more confidence and trust in courts and the legal system, Congress, the executive branch, and the Supreme Court relative to natives – all to a large and statistically significant extent (Nowrasteh & Powell, 2021, pp. 174-7). The pessimistic interpretation is that immigrants just haven't had enough experience with those institutions to efficiently distrust them, but that also means U.S. institutions compare favorably with institutions in their countries of origin. No surprise there. More importantly, immigrants have much more trust in major private companies than natives do to a statistically significant extent. On a scale of 0-10, where 0 is “no trust” and 10 is “complete trust,” 43 percent of immigrants rate their level of trust on a scale of 6 or higher compared to just 31 percent of natives.
Trust and formal institutions may substitute for each other (Cook et al., 2005; Nowrasteh & Powell, 2021, p. 172-4). In China, private sector guanxi networks help facilitate business relationships in the absence of good formal institutions (Xin & Pearce, 1996). Alesina and La Ferrara (2002) found that Americans living in racially fragmented communities have less general social trust but trust formal institutions. Trust and formal institutions also seem to substitute in financial transactions (Cline & Williamson, 2016). Macro-level evidence is consistent with the theory that generalized social trust is most closely related to increased income when formal institutions are weak (Aherlup et al., 2009).
The Implications of Trust for Immigration Policy
Assuming that my criticisms above are overstated or wrong, that trust really is an important mechanism, and that immigrants and their descendants don’t assimilate, that implies that U.S. immigration policy should still be less restrictive than it is today. The WVS over the 2005-2022 waves reports that the "can trust" response in the United States is an average of about five percentage points higher than the same response to the same trust question in the GSS — a small difference. The U.S. GSS reports that 32 percent of Americans said they "can trust" in the 2017-2022 period, less than the 37 percent reported by the WVS for the same period – again, a small difference. In the same period, the WVS also reported that 27 percent of global responses were "can trust" – a small difference compared with the United States, especially the larger sample GSS. Of course, responses between countries vary a lot.
The population of countries with a higher “can trust” response than the United States in the WVS 2017-2022 wave is 1.75 billion, and that’s from a survey of only 88 countries. The population of countries with “can trust” responses in the WVS survey that are higher than the native-born U.S. GSS measure is 1.9 billion. An immigration policy to admit people from high trusting countries based solely on the WVS data would include China, which is consistent with Jones’s writing elsewhere in his book, but would oddly exclude France, South Korea, Japan, and Singapore. Using the GSS for U.S. responses would include South Korea, Japan, and Singapore – but just barely. Sorry France. A labor mobility zone for the United States that includes all countries with higher trust would be the most radical international labor market deregulation in history, 3-4 times larger than Schengen. Politically, we can’t even get a free movement agreement with Canada.
Trust: What’s left?
If you still think that trust matters, there is one low-cost way to increase trust: stop lying and punish liars. Few things destroy trust more than lying, especially when liars get away with it. So, the government should stop lying, there should be an easier legal mechanism to remove politicians and bureaucrats who lie, and public officials should abandon the idea of “noble lies.” The amount of nonsense and lies promulgated by government officials and politicians during the Vietnam War, the COVID-19 pandemic, the War on Terrorism, and other events of the last 60 years have diminished public trust in the government. Daniel Ellsberg, the famous whistleblower who released the Pentagon Papers during the Nixon administration, said:
The public is lied to every day by the President, by his spokespeople, by his officers. If you can't handle the thought that the President lies to the public for all kinds of reasons, you couldn't stay in the government at that level, or you're made aware of it, a week . . . The fact is Presidents rarely say the whole truth—essentially, never say the whole truth—of what they expect and what they're doing and what they believe and why they're doing it and rarely refrain from lying, actually, about these matters.
That behavior must have influenced trust, especially trust in the government which might impact generalized trust (Brehm & Rahn, 1997; Bargsted et al., 2022). Whistleblower protections are weak. Of course, there are plenty of other good reasons to distrust the government, but constant lying increases it. Still, a government that has earned trust is likely to be better than one that doesn't, other things being equal. Private institutions and organizations, even the media, better police lies and punish liars than the government. Still, they should also do more to punish liars and should unceremoniously, brutally, and publicly fire them before they resign. But low trust is not a good reason to keep the borders closed and leave tens of trillions of dollars on the sidewalk.
The trust-growth literature is a slender reed that bears great weight in Jones’s book. Jones presents trust as a sturdier reed by ignoring problems that are widely and extensively noted in the peer-reviewed academic literature. Jones’s book is already light on theory and mechanisms. Removing or reducing trust as a plausible mechanism weakens it even further. Even so, liberalizing immigration between higher-trust countries would be a radical liberalization.
Stay tuned for the next installment.
References
Akcomak, S. and Weel, B. (2009). “Social Capital, Innovation and Growth: Evidence from Europe. European Economic Review, 53 (5), 544–67.
Aherlup, P., Olsson, O., and Yanagizawa-Drott, D. (2009). “Social Capital vs. Institutions in the Growth Process.” European Journal of Political Economy, 25(1), 1-14.
Algan, Y. and Cahuc, P. (2010). “Inherited Trust and Growth.” American Economic Review, 100 (5), 2060– 92.
Algan, Y. and Cahuc, P. (2013). “Trust and Growth.” Annual Review of Economics, 5 (1), 521–49.
Alesina, A. and La Ferrara, E. (2002). “Who Trusts Others?” Journal of Public Economics, 85(2), 207-34.
Beugelsdijk, S. and Maseland, R. (2011). Culture in Economics: History, Methodological Reflections, and Contemporary Applications. Cambridge University Press.
Bargsted, M., Ortiz, C., Cáceres, I., Somma, N.M. (2022). “Social and Political Trust in a Low Trust Society.” Political Behavior.
Bjørnskov, C. (2009). “Social Trust and the Growth of Schooling.” Economics of Education Review, 28 (2), 249–257.
Bjørnskov, C. (2018). “Social Trust and Economic Growth.” In E. Uslaner (Ed.), The Oxford Handbook of Social and Political Trust.
Blume, L. E., Brock, W. A., Durlauf, S. N. and Ioannides, Y. M. (2010). “Identification of Social Interactions,” (Economics Series No. 260). Institute for Advanced Studies.
Blundell, R. (2017). “What Have We Learned from Structural Models?” American Economic Review, 107(5), 287-92.
Bohnet, I., Greig, F., Herrmann, B. and Zeckhauser, R. (2008). “Betrayal Aversion: Evidence from Brazil, China, Oman, Switzerland, Turkey, and the United States.” American Economic Review, 98 (1), 294–310.
Brehm, J. and Rahm, W.M. (1997). “Individual-Level Evidence for the Causes and Consequences of Social Capital.” American Journal of Political Science, 41: 999-1023.
Briggeman, J. (2022). “Long-Run Determinants of Economic Growth: Putterman and Weil Revisited.” Econ Journal Watch, 19(1), 85-108.
Butler, J. V., Giuliano, P., & Guiso, L. (2016). “The Right Amount of Trust.” Journal of the European Economic Association, 14(5), 1155–80.
Cachanosky, N., Padilla, A., and Gómez, A. (2021). “Immigration and Institutional Change: Did Mass Immigration Cause Peronism in Argentina?” Journal of Economic Behavior and Organization, vol. 184(C), 1-15.
Caplan, B. (2022). “The Case for 50 Percent Open Borders.” Reason, November 15, 2022.
Clemens, M.A. (2011). “Economics and Emigration: Trillion-Dollar Bills on the Sidewalk?” Journal of Economic Perspectives, 25(3), 83-106.
Clemens, M.A. & Pritchett, L. (2019). “The New Economic Case for Migration Restrictions: An Assessment,” Journal of Development Economics, 138(C), 153-164.
Cline, B. and Williamson, C. (2016). “Trust and Regulation of Corporate Self-Dealing.” Journal of Corporate Finance, 41, 572-90.
Cook, K.S., Hardin, R., and Levi, M. (2005). Cooperation Without Trust? Russel Sage Foundation.
Dincer, O. and Uslaner, E. (2010). “Trust and Growth.” Public Choice, 142 (59), 60–61.
Dippel, C. and Heblich, S. (2021). “Leadership in Social Movements: Evidence from the ‘Forty-Eighters’ in the Civil War.” American Economic Review, 111(2), 472-505.
Duncan, B. and Trejo, S. (2016). “The Complexity of Immigrant Generations: Implications for Assessing the Socioeconomic Integration of Hispanics and Asians.” Industrial and Labor Relations Review, 70 (5), 1146–75.
Durlauf, S. (2002). “On the Empirics of Social Capital.” The Economic Journal, 112 (483), 459–79.
Durlauf, S., Johnson, P. and Temple, J. (2005). “Growth Econometrics.” In S. Durlauf and P. Aghion (Eds.), Handbook of Economic Growth.
Eder, C. (2018). “A Replication of ‘Inherited Trust and Growth.’” Economics Bulletin, 38(1), 574-82.
Ermisch, J., Gambetta, D., Laurie, H., Siedler, T. and Uhrig, S. (2009). “Measuring People’s Trust.” Journal of the Royal Statistical Society, 172 (4), 749–69.
Fehr, E., Fischbach, U., Rosenbladt, B., Schupp, J. and Wagener, G. (2002). “A Nation-Wide Laboratory: Examining Trust and Trustworthiness by Integrating Behavioral Experiments into Representative Surveys.” Schmollers Jahrbuch, 122 (4), 519–42.
Felten, E. (2000). “Finders Keepers?” Reader’s Digest, 158 (948).
Fernandez, R. (2011). “Does Culture Matter?” In J. Benhabib, A. Bisin and M. Jackson (Eds.), Handbook of Social Economics, volume 1a.
Fulford, S.L., Petkov, I., Schiantarelli, F. (2020). “Does It Matter Where You Came From? Ancestry Composition and Economic Performance of U.S. Counties, 1850-2010." Journal of Economic Growth, 25, 341-80.
Gächter, S., Herrmann, B. and Thöni, C. (2004). “Trust, Voluntary Cooperation, and Socioeconomic Background: Survey and Experimental Evidence.” Journal of Economic Behavior & Organization, 55(4), 505–531.
Gambetta, D. (2000). “Can We Trust Trust? In D. Gambetta (Ed.), Trust: Making and Breaking Cooperative Relations (pp. 213–237). Blackwell.
Giavazzi, F., Petkov, I. and Schiantarelli, F. (2019). “Culture: Persistence and Evolution.” Journal of Economic Growth, 24 (2), 117–54.
Glaeser, E., Liabson, D. and Sacerdote, B. (2002). “An Economic Approach to Social Capital.” The Economic Journal, 112, 437–58.
Glaeser, E., Laibson, D., Scheinkman, J. and Soutter, C. (2000). “Measuring Trust.” Quarterly Journal of Economics, 115 (3), 811–46.
Guiso, L., Sapienza, P. and Zingales, L. (2009). “Does Culture Affect Economic Outcomes?” Journal of Economic Perspectives, 20 (2), 23–48.
Guiso, L., Sapienza, P. and Zingales, L. (2011). “Civic Capital as the Missing Link.” In J. Benhabib, A. Bisin and M. Jackson (Eds.), Handbook of social economics, volume 1a.
Hardin, R. (2006). Trust. Cambridge University Press.
Holm, H. and Danielson, A. (2005). “Tropic Trust Versus Nordic Trust: Experimental Evidence from Tanzania and Sweden.” The Economic Journal, 115(503), 505–32.
Ikeda, S. (2008). “The Meaning of ‘Social Capital’ as It Relates to the Market Process.” i, 21 (2), 167–82.
Jones, G. (2016). Hive Mind: How Your Nation’s IQ Matters So Much More than Your Own. Stanford University Press.
Jones, G. (2018). “The Populist Revolts of 2016: A Hidden Cost of Immigration.” Independent Review, 22(3), 341-47.
Jones, G. (2020). 10% Less Democracy. Stanford University Press.
Jones, G. (2023). The Culture Transplant: How Migrants Make the Economics They Move to A Lot Like the Ones They Left. Stanford University Press.
Karlan, D. (2005). “Using Experimental Economics to Measure Social Capital and Predict Financial Decisions.” American Economic Review, 95 (5), 1688–99.
Kelly, M. (2019). “The Standard Errors of Persistence.” UCD Centre for Economic Research Working Paper Series, No. WP19/13.
Knack, S. and Keefer, P. (1997). “Does Social Capital Have an Economic Payoff? A Cross-Country Investigation.” The Quarterly Journal of Economics, 112 (4), 1251–88.
Lucas, R.E. (1988). “On the Mechanics of Economic Development.” Journal of Monetary Economics, 22(1), 3-42.
Makrychoriti, P., Pasiouras, F., & Tasiou, M. (2021). “Financial Stress and Economic Growth: The Moderating Role of Trust.” Kyklos, 75(1), 48-74.
Martinez, L.R. (2022). “How Much Should We Trust the Dictator’s GDP Growth Estimates?” Journal of Political Economy, 130(10), 2731-69.
Miller, A. and Mitamura, T. (2003). “Are Surveys on Trust Trustworthy?” Social Psychology Quarterly, 66 (1), 62–70.
Moore, M. (1999). “Truth, Trust and Market Transactions: What Do We Know?” Journal of Development Studies, 36 (1), 74–88.
Müller, D., Torgler, B. and Uslaner, E. (2012). “A Comment on ‘Inherited Trust and Growth.’” Economics Bulletin, 32 (2), 1481–88.
Näf, M. and Schupp, J. (2009). “Can we Trust the Trust Game? A Comprehensive Examination.”
Nannestad, P. (2008). “What Have We Learned About Generalized Trust, If Anything?” The Annual Review of Political Science, 11: 413-36.
Nowrasteh, A. (2018). “Singapore’s Immigration System: Past, Present, and Future.” Cato Working Paper 53.
Nowrasteh, A. and Forrester, A. (2020). "Trust Doesn't Explain Regional U.S. Economic Development and Five Other Theoretical and Empirical Problems with the Trust Literature.” Cato Institute Working Paper 57, https://www.cato.org/sites/cato.org/files/2020-01/working-paper-57-updated.pdf.
Nowrasteh, A. and Powell, B. (2021). Wretched Refuse? The Political Economy of Immigration and Institutions. Cambridge University Press.
Owyang, M.T. and Shell, H. (2017). “China’s Economic Data: An Accurate Reflection, or Just Smoke and Mirrors?” The Regional Economist, St. Louis Federal Reserve Bank, July 25, 2017.
Putnam, R. D. (2000). Bowling Alone: The Collapse and Revival of American Community. Simon & Schuster.
Putterman, L. and Weil, D. (2010). “Post-1500 Population Flows and the Long-Run Determinants of Economic Growth and Inequality.” Quarterly Journal of Economics, 125(4), 1627-82.
Romer, P. (1986). “Increasing Returns and Long-Run Growth.” Journal of Political Economy, 94(5), 1002- 37.
Rose, D. C. (2011). The Moral Foundation of Economic Behavior. Oxford, UK: Oxford University Press.
Rose, D. C. (2019). Why Culture Matters Most. Oxford, UK: Oxford University Press.
Roth, F. (2009). “Does Too Much Trust Hamper Growth?” Kyklos, 62 (1), 103–28.
Sapienza, P., Toldra-Simats, A. and Zingales, L. (2013). “Understanding Trust.” The Economic Journal, 128 (873), 1313–32.
Tabellini, G. (2008). “Institutions and Culture.” Journal of the European Economic Association, 6 (2-3), 255–94.
Tabellini, G. (2010). “Culture and Institutions: Economic Development in the Regions of Europe.” Journal of the European Economic Association, 8 (4), 677–716.
Torsvik, G. (2000). “Social Capital and Economic Development: A Plea for the Mechanisms.” Rationality and Society, 12(4), 451–76
Uslaner, E.M. (2008a). “The Foundations of Trust: Macro and Micro.” Cambridge Journal of Economics, 32 (2), 289–94
Uslaner, E.M. (2008b). “Where You Stand Depends upon Where Your Grandparents Sat: The Inheritability of Generalized Trust.” The Public Opinion Quarterly, 72(4), 725-40.
Weil, D. (2005). Economic Growth. Addison-Wesley
Xin, K. and Pearce, J.L. (1996). “Connections as Substitutes for Formal Institutional Support.” The Academy of Management Journal, 39(6), 1641-58.
Zak, P. and Knack, S. (2001). “Trust and Growth.” The Economic Journal, 111 (470), 295–321.
[i] Some of this section is taken from our submitted R&R that is currently with an economics journal.
Really excellent job! One of the best articles I've read this year. Very thorough takedown.
Looking forward to the next installment