Using de-identified Facebook data on roughly 17 million social ties in 55 countries, Gee, Jones, Fariss, Burke, and Fowler (2017) documented a cross-national 'paradox of weak ties': most people find jobs through one of their numerous weak ties, yet within every country the probability that a specific friend transmits a job increases, rather than decreases, with the strength of the tie to that friend. They further showed that the marginal value of a strong tie is positively correlated across countries with income inequality as measured by the Gini coefficient, but they could not identify the direction of causation or the micro-mechanisms involved. The open problem is to establish the causal structure and mechanisms behind this inequality–strong-tie-value association, contending with the measurement limitations (a noisy 'sequential job' proxy with roughly 20–30 percent dyad-level survey-validated precision and a Facebook-user sample that is not nationally representative) that constrain the original observational design.

Background

Since Granovetter's classic study, social network research has distinguished weak ties (acquaintances contacted infrequently) from strong ties (close friends contacted often), and asked which kind helps people find jobs (Granovetter, 1973, 'The Strength of Weak Ties'). Granovetter argued that weak ties are collectively valuable because they bridge distant parts of a network and carry novel information. A pair of papers using Facebook data complicated this picture. Gee, Jones, and Burke (2017) analyzed US data and found that although most people who found a job through a friend did so via one of their many weak ties, each individual strong tie was more likely to transmit a job than each individual weak tie: weak ties dominate in aggregate only because they are far more numerous. Gee, Jones, Fariss, Burke, and Fowler (2017) then applied a single measurement and estimation framework across 55 countries. They operationalized job transmission through a 'sequential job' dyad indicator: both members of a friendship list the same employer, the friend started there at least one year before the user, and the friendship predates the user's start by at least one year. Tie strength for each user-friend pair was measured, before the job started, by the fraction of the user's photo tags and wall posts directed to that friend and by the Jaccard index of mutual friends (the size of the intersection of the two friends' lists divided by the size of their union). Within-user fixed-effects regressions, which compare a user's different friends and so hold constant all user-level attributes, showed that in every one of the 55 countries a stronger tie is more likely to transmit a job, with substantial variation across countries in the size of this marginal effect; a reversed-timing placebo test (re-estimating the transmission indicator with the friend's and user's job-start timing reversed, which should reproduce the original signal if it reflected pure dyad-level confounding rather than genuine transmission) suggested the association is not purely dyad-level confounding. Strikingly, the country-level marginal value of a strong tie is positively correlated with national income inequality as measured by the Gini coefficient (a standard summary statistic of income inequality ranging from 0 for perfect equality to 1 for maximal inequality), robust to uncertainty quantified by a parametric bootstrap (repeated re-estimation on simulated datasets drawn from a fitted model) and to controls for mean income and population. The authors propose, but do not test, candidate mechanisms: lower generalized trust (trust in people in general rather than in one's known contacts) in unequal societies may make strong ties more useful in job finding; and socioeconomically bounded strong-tie networks may lock people into their current income level and perpetuate inequality. Because the design is observational and cross-sectional, the direction of causation between inequality and strong-tie value is unidentified and the mechanisms are untested conjectures; the sequential-job proxy's dyad-level precision (validation surveys correctly identified a genuinely helpful tie between about 20 and 30 percent of the time, though the authors report no evidence that this noise biases their results) and the platform-based sample, which is not nationally representative, constrain external validity. Later work has clarified pieces of the surrounding landscape: a large-scale LinkedIn experiment causally confirmed that per-tie network composition affects job mobility, with nonlinear effects and strong ties more useful in less digitalized industries (Rajkumar, Saint-Jacques, Bojinov, Brynjolfsson, and Aral, 2022); Facebook-based studies of 'economic connectedness' — friendship links across socioeconomic lines — linked cross-class ties to upward income mobility and decomposed their absence into exposure and friending bias (Chetty et al., 2022); and reviews of the strength-of-weak-ties literature still list the marginal-effect and causality questions as open frontiers (Kim and Fernandez, 2023). None of this work identifies the causal role of inequality itself.

Problem Statement

The problem is to explain the inequality-strong-tie association documented by Gee, Jones, Fariss, Burke, and Fowler (2017): why is the marginal value of a strong tie in job transmission larger in countries with greater income inequality? The open target is the causal structure and the micro-mechanisms of this association. Two limitations of the original design form the background any new evidence must contend with rather than assume away: the 'sequential job' proxy was survey-validated at roughly 20-30 percent dyad-level precision (the authors report no evidence that this noise biases their results, and a direct analysis of the survey data yielded statistically significant and substantively stronger findings), and the Facebook-based sample is not nationally representative. The causal structure and micro-mechanisms to be established are: (1) Causal structure: determine whether greater income inequality increases the marginal effectiveness of strong ties in job transmission, whether the stronger role of strong ties instead causes or perpetuates inequality, whether both are driven by common factors, or some mixture of these; observational and correlational evidence, however suggestive, does not suffice. (2) Mechanisms: identify the micro-mechanisms underlying the inequality–strong-tie-value association, whichever causal structure (1) establishes, and empirically test the two candidate channels proposed in the source work - lower generalized trust in more unequal countries making trusted strong ties more useful in job finding, and socioeconomically bounded strong-tie networks that lock people into their current income level and perpetuate inequality - rather than leaving them as competing hypotheses. The source authors suggest that the same collective-import-of-weak-ties-but-greater-marginal-value-of-strong-ties pattern may recur in other settings (team productivity, disease contagion, the proliferation of trends, and marriage).

References

  1. Granovetter, M. S. (1973). The Strength of Weak Ties. American Journal of Sociology, 78(6), 1360-1380. https://doi.org/10.1086/225469
  2. Gee, L. K., Jones, J. J., & Burke, M. (2017). Social Networks and Labor Markets: How Strong Ties Relate to Job Finding on Facebook's Social Network. Journal of Labor Economics, 35(2), 485-518. https://doi.org/10.1086/686225
  3. Gee, L. K., Jones, J. J., Fariss, C. J., Burke, M., & Fowler, J. H. (2017). The paradox of weak ties in 55 countries. Journal of Economic Behavior & Organization, 133, 362-372. https://doi.org/10.1016/j.jebo.2016.12.004
  4. Rajkumar, K., Saint-Jacques, G., Bojinov, I., Brynjolfsson, E., & Aral, S. (2022). A causal test of the strength of weak ties. Science, 377(6612), 1304-1310. https://doi.org/10.1126/science.abl4476
  5. Kim, M., & Fernandez, R. M. (2023). What Makes Weak Ties Strong? Annual Review of Sociology, 49(1), 177-193. https://doi.org/10.1146/annurev-soc-030921-034152
  6. Neal, Z. P. (2024). The not-so-forbidden triad: Evaluating the assumptions of the Strength of Weak Ties. Network Science, 12(3), 289-304. https://doi.org/10.1017/nws.2024.12
  7. Ruiz-Palazuelos, S., Espinosa, M. P., & Kovářík, J. (2023). The weakness of common job contacts. European Economic Review, 160, 104594. https://doi.org/10.1016/j.euroecorev.2023.104594
  8. Chetty, R., Jackson, M. O., Kuchler, T., Stroebel, J., Hendren, N., Fluegge, R. B., Gong, S., Gonzalez, F., Grondin, A., Jacob, M., Johnston, D., Koenen, M., Laguna-Muggenburg, E., Mudekereza, F., Rutter, T., Thor, N., Townsend, W., Zhang, R., Bailey, M., Barberá, P., Bhole, M., & Wernerfelt, N. (2022). Social capital I: measurement and associations with economic mobility. Nature, 608, 108-121. https://doi.org/10.1038/s41586-022-04996-4
  9. Chetty, R., Jackson, M. O., Kuchler, T., Stroebel, J., Hendren, N., Fluegge, R. B., Gong, S., Gonzalez, F., Grondin, A., Jacob, M., Johnston, D., Koenen, M., Laguna-Muggenburg, E., Mudekereza, F., Rutter, T., Thor, N., Townsend, W., Zhang, R., Bailey, M., Barberá, P., Bhole, M., & Wernerfelt, N. (2022). Social capital II: determinants of economic connectedness. Nature, 608, 122-134. https://doi.org/10.1038/s41586-022-04997-3
  10. Bühler, J., Häner-Müller, M., & Schaltegger, C. A. (2026). Who You Meet Matters: International Evidence on Economic Connectedness and Social Mobility. The B.E. Journal of Economic Analysis & Policy. https://doi.org/10.1515/bejeap-2025-0497
  11. Miller, C., & Schmutte, I. M. (2021). The Dynamics of Referral Hiring and Racial Inequality: Evidence from Brazil. NBER Working Paper 29246. https://doi.org/10.3386/w29246
  12. Armona, L. (2025). Online Social Network Effects in Labor Markets: Evidence from Facebook's Entry to College Campuses. Review of Economics and Statistics, 107(4), 873-888. https://doi.org/10.1162/rest_a_01354
  13. Barwick, P. J., Liu, Y., Patacchini, E., & Wu, Q. (2023). Information, Mobile Communication, and Referral Effects. American Economic Review, 113(5), 1170-1207. https://doi.org/10.1257/aer.20200187
  14. Eckles, D., Mossel, E., Rahimian, M. A., & Sen, S. (2024). Long ties accelerate noisy threshold-based contagions. Nature Human Behaviour, 8(6), 1057-1064. https://doi.org/10.1038/s41562-024-01865-0
  15. Jahani, E., Fraiberger, S. P., Bailey, M., & Eckles, D. (2023). Long ties, disruptive life events, and economic prosperity. Proceedings of the National Academy of Sciences, 120(28), e2211062120. https://doi.org/10.1073/pnas.2211062120