Information, Mobile Communication, and Referral Effects: Barwick, Liu, Patacchini & Wu (2023)

What this is. The paper’s core results, the identification strategy, and the estimating equations: enough to know what it found and how, without reading all 38 pages. To replicate or extend, read the full source at the original.

TL;DR

The paper exploits geocoded cellphone records from a major Chinese telecom provider to study how social contacts relate to job search and hiring. It documents (i) an inverted U-shaped spike in call frequency between switchers and their referrers that peaks just before a job change, with no corresponding pattern for non-referrer friends; (ii) a referral effect of 0.35 on job location choice - having a social contact working at a location nearly triples the probability of switching there; and (iii) referral jobs are associated with better worker outcomes: higher pay, shorter commutes, and more full-time transitions. Referral hiring is also associated with faster firm growth in descriptive regressions. The study cannot distinguish whether information reaches the worker or the employer. Effect heterogeneity shows referrals matter especially when information asymmetry is more severe, as for young workers, rural-to-urban movers, and sector-changers. Topa (2001) provides the foundational social-interactions framework; Bayer, Ross, and Topa (2008) introduced the residential-neighbor proxy approach that this paper extends using direct communication data.

Core results

Magnitudes and significance are as reported; */**/*** = 10%/5%/1%. Locators point into the source PDF.

#	Result	Locator	Magnitude
R1	Referral effect on job location choice: having a friend at location l increases probability of switching there by 0.35	Table 3, col 2, p. 1186; text p. 1185	Coeff = 0.35 (SE 0.01); mean baseline probability = 0.09; N = 915,251 switcher-location pairs
R2	Inverted U-shape in referrer call frequency: calls between switchers and referrers peak just before the job switch; nonreferrer calls are flat throughout	Figure 3, p. 1188	Referrer-pair coefficient peaks at approx +8 above baseline just before the switch; nonreferrer pairs approx 0 throughout; 238,092 referrer-month obs vs 4,759,176 nonreferrer obs
R3	Referral effect amplified for high information-asymmetry groups: rural-to-urban movers and sector-changers show substantially larger referral effects	Table 4, cols 5-6, p. 1191	Friend x rural-to-urban = +0.32 (SE 0.05); Friend x changing sector = +0.21 (SE 0.02); baseline Friend 0.34 (col 5) / 0.32 (col 6)
R4	Referral wage premium: referral jobs pay RMB 620 more per year, about 2 percent above the mean wage	Table 7, col 1, p. 1199; text p. 1198	0.62 thousand RMB (SE 0.31); mean wage approx 31 thousand RMB/year; N = 17,615
R5	Referral jobs are more likely to be full-time: having a referrer at the new workplace raises the probability of a part-time to full-time transition by 1.4 percentage points	Table 7, col 3, p. 1199; text p. 1198	0.014 (SE 0.007); approx 2% relative increase; N = 19,431
R6	Referral jobs have shorter commutes: referral raises probability of a shorter commute by 9 percentage points	Table 7, col 4, p. 1199; text p. 1198	0.09 (SE 0.01); approx one-third of job changes involve a shorter commute; N = 29,117
R7	Firm benefit: net labor inflow: firms hiring through referrals gain 63 percent more workers (log net inflow) in the most saturated specification	Table 8, Panel A col 4, p. 1201	gamma = 0.63 (SE 0.14); R-squared = 0.66
R8	Firm benefit: matching rate: firms hiring through referrals have an 84 percent higher matching rate (log net inflow over vacancies)	Table 8, Panel B col 8, p. 1201	gamma = 0.84 (SE 0.27); average matching rate = 1.53 for large firms
R9	Informal referrals are more common in China than the US: respondents reported finding jobs through friends more often in China	Figure 1, p. 1178	Approximately 38% in China and 28% in the US, read from the plotted bars
R10	Sample demographics broadly resemble the national cellphone-user survey: sample and CFPS values are close for age and province of birth, with a lower female share in the paper’s sample	Table 1, p. 1179	Female: 0.36 vs 0.45; born in local province: 0.75 vs 0.76; age midpoint: 40.18 vs 39.28 years
R11	Social-contact counts do not spike before job changes: total weekly contacts are stable, supporting the use of pre-existing ties	Figure 2, p. 1180	Average is 23-25 contacts per week; no pre-switch spike reported
R12	Communication predicts neighborhood worker flows	Table 2, Panel A, p. 1182	Total calls coefficient = 0.101 (SE 0.016); R² rises from 0.037 to 0.170; inverse-hyperbolic-sine result: doubling calls is associated with 16% more flows (text p. 1182)
R13	Calls improve out-of-sample prediction of worker flows	Table 2, Panel B, p. 1182	RMSE falls from 0.253 to 0.181; MAPE falls from 1.162% to 1.037%
R14	Moved-away friends are a placebo for current referrals: their location-choice association is smaller than for current referrers	Table 3, col 3, p. 1186	Moved-away friend coefficient = 0.07 (SE 0.03), versus current Friend = 0.34 (SE 0.01)
R15	Friends who live but do not work at the destination have a smaller association	Table 3, col 4, p. 1186	Coefficient = 0.15 (SE 0.02), versus current Friend = 0.34 (SE 0.01)
R16	Second-degree friends have a smaller association than direct referrers	Table 3, col 5, p. 1186	Friends-of-friends coefficient = 0.14 (SE 0.01), versus direct Friend = 0.33 (SE 0.01)
R17	Placebo-friend communication lacks the referrer event pattern: calls to moved-away and locally resident nonworking friends do not show a systematic pre-switch peak	Figure 4, pp. 1189-1190	No systematic peak reported for 29,148 moved-away-friend months or 62,472 local nonworking-friend months; referrer sample = 238,092 observations
R18	Referral effects increase with distance between old and new workplaces	Table 4, col 2, p. 1191	Friend × distance (old job to new job) = 0.002 (SE 0.0004)
R19	Referral effects increase with distance between home and new workplace	Table 4, col 3, p. 1191	Friend × distance (home to new job) = 0.002 (SE 0.0003)
R20	Referral effects are larger for young workers	Table 4, col 4, p. 1191	Friend × young (ages 25-34) = 0.04 (SE 0.01)
R21	Residential neighbors are a useful but attenuated referral proxy	Table 5, cols 1 and 3, p. 1193	Neighbor coefficient = 0.21 (SE 0.01) alone; 0.18 (SE 0.01) with Friend, not neighbor coefficient = 0.25 (SE 0.01)
R22	Same-birth-county coworkers are a weaker proxy than direct communication ties	Table 5, cols 2 and 4, p. 1193	Same-birth-county coefficient = 0.10 (SE 0.01) alone and 0.09 (SE 0.01) with Friend, not same birth county = 0.35 (SE 0.03)
R23	Observed communication is more frequent among neighbor and coworker ties	Text p. 1193 (online Appendix Table S8)	Same-location neighbors make 4.5 times as many calls as random pairs in the same neighborhood; coworkers make four times as many calls as pairs working in the same neighborhood
R24	Stronger communication ties have larger referral effects	Text pp. 1193-1194 (online Appendix Table S9)	A one-standard-deviation increase in the call share is associated with a 6% increase in the referral effect; SD of call share = 0.06
R25	More diverse social contacts are associated with more referral job changes	Text p. 1194 (online Appendix Table S10)	One-standard-deviation increases in social entropy and income entropy are associated with 6% and 3% increases, respectively, in the probability of using referrals
R26	The referral estimate survives explicit homophily controls	Table 6, cols 1-2, p. 1195	Friend coefficient = 0.33 (SE 0.03) with observed homophily controls and 0.34 (SE 0.01) with 100 k-means clusters
R27	Preference for working with friends is small relative to the referral association	Table 6, col 3, p. 1195	More friends at the new than old workplace coefficient = 0.002 (SE 0.001), compared with baseline referral coefficient = 0.35 (SE 0.01)
R28	The referral result is robust to alternative pre-switch network cutoffs	Text pp. 1196-1197 (online Appendix Table S11)	The paper reports robustness for contact cutoffs from one to five months before switching; coefficients are not stated in the main-text discussion
R29	Alternative friend definitions yield similar referral estimates	Text p. 1197 (online Appendix Table S12)	Friend locations observed for at least four weeks: 0.36; reciprocal-call friends with that coverage rule: 0.38; baseline = 0.35
R30	Friend-coverage split does not materially change the referral estimate	Text p. 1197 (online Appendix Table S12, cols 3-4)	Above- versus below-median coverage estimates differ modestly and not significantly; median coverage cutoff = 48%
R31	Referral jobs have higher coworker housing values	Table 7, col 2, p. 1199; text p. 1198	Change in coworker housing price = 0.07 thousand RMB/m² (SE 0.04), about 0.5% of the city average
R32	Referrals increase transitions from non-SOE to SOE jobs	Table 7, col 5, p. 1199; text p. 1198	Coefficient = 0.012 (SE 0.005), an 11% increase over the 0.11 mean
R33	Referrals are associated with faster firm growth	Table 8, Panel C, col 12, p. 1201	Coefficient = 0.45 (SE 0.11); paper reports a 45% increase; median large-firm growth rate = 4%
R34	Communication channels are complements in the external validation evidence	Text p. 1202 (online Appendix Tables S17-S18)	Call volume is positively and significantly correlated with 4G compatibility, internet allowance, and browsing; users who call more also text and use WeChat more; exact coefficients are not stated in the main text
R35	Referrer communication rises during unemployment and peaks before reemployment	Figure 5, p. 1204	Inverted-U pattern during job search; 53,244 referrer-month and 747,804 nonreferrer-month observations
R36	The referral association also appears for workers reemployed after a gap	Text p. 1204 (online Appendix Table S21)	Referral effect ranges from 0.31 to 0.33; 1,677 of 3,638 identified workers found jobs through referrals
R37	The referral association is present in the unrestricted switcher sample	Table 3, col 1, p. 1186	Friend coefficient = 0.34 (SE 0.01); N = 1,120,797 switcher-location pairs
R38	The baseline referral estimate is stable with extensive local labor-market controls	Table 3, col 6, p. 1186	Friend coefficient = 0.33 (SE 0.01); N = 915,251; 16,468 old-by-new neighborhood-pair fixed effects

Overall (paper’s conclusion). Information provided by social contacts mitigates information asymmetry in labor markets and facilitates better worker-firm matching. The inverted U-shape in referrer communication around job changes, absent for non-referrer friends, provides evidence of job-related information exchange within referral pairs and weighs against homophily and sorting as the sole explanation. The study cannot distinguish whether information is shared with workers or employers. Referred employees have higher wages, shorter commutes, and more full-time transitions; in descriptive firm regressions, referral hiring is associated with faster growth and higher matching rates.

Theory / model

This paper has no formal economic model. It tests three linked hypotheses derived from the theoretical literature on information transmission in labor markets (Topa 2001):

1. Information channel hypothesis. Referral pairs exchange job-relevant information, either reaching job seekers or informing employers about candidates, generating an increase in communication intensity in the months before the job change. The prediction is an inverted U-shape in call frequency between referrer pairs centered on the event month, with no such spike for non-referrer friends.

2. Referral effect hypothesis. Having a social contact working at a given location raises the probability of switching there. The coefficient captures both information provision (the referrer informs the job seeker of an opening) and endorsement (the referrer vouches for the candidate to the employer). Under either channel, the referral should increase location choice probability.

3. Information-asymmetry amplification. The referral effect should be larger when information between workers and firms is more asymmetric: for young workers with limited labor market experience, rural-to-urban movers unfamiliar with urban job markets, and sector-changers whose skills are less observable. If the mechanism is pure preference (working near friends), no such heterogeneity would be predicted.

Identification strategy. The key threats are homophily (friends share unobserved location preferences) and sorting (friends cluster in locations with unobserved job opportunities). The paper addresses these via:

• Origin-destination neighborhood-pair fixed effects (equation 1, p. 1184): the referral coefficient beta is identified from within-pair variation, comparing job switchers who move between the same old-new neighborhood pair but have different social networks. This controls for all aggregate pair-specific attributes including industry composition, labor demand, and amenities.

• Falsification tests using friend type (Table 3, cols 3-4, p. 1186): friends who recently moved away from location l have a coefficient of 0.07 (far below the 0.35 for current referrers). Friends who live in the new location’s neighborhood but do not work there have a coefficient of 0.15. If homophily or the specific neighborhood drove the result, these coefficients would be similar to the baseline. Current employment at the destination is what matters, consistent with information about job openings being the active ingredient.

• Vacancy restriction (Table 3, col 2, p. 1186): the baseline sample restricts to switchers with at least one alternative location in the same neighborhood offering the same occupation and salary range, ruling out the concern that the friend dummy proxies for the only available matching job.

• Homophily controls (Table 6, p. 1195): adding same-gender, same-age-group, same-birth-county, same-housing-price controls and k-means cluster dummies for switcher-friend pairs leaves the estimate stable at 0.33-0.34 in specifications that account for measured similarity and clusters.

The paper also compares its call-based referral measure with proxy-based approaches common in the literature: residential neighbors in the spirit of Bayer, Ross, and Topa (2008) (coefficient 0.21) and same-birth-county coworkers (0.10). The call-based measure exceeds both proxies by a margin that is statistically significant at the one-percent level (Table 5, cols. 3-4, p. 1193).

Method

The paper applies OLS panel regression with fixed effects and an event study. The builds on panel-regression and event-study technique primitives. The four estimating equations are introduced on pp. 1181-1184 and 1197-1200.

Preliminary correlation (Table 2, p. 1182). The relationship between information flow (call volume) and worker flows across neighborhood pairs is established via OLS with origin and destination fixed effects. Adding call volume as a regressor raises the R-squared from 0.037 to 0.17, and doubling call volume is associated with a 16 percent increase in worker flows (inverse-hyperbolic-sine specification). This motivates using communication intensity as a proxy for information provision. The unnumbered neighborhood-pair regression in Table 2 is written out here from its reported variables (p. 1182); columns 1-3 use the listed subset of regressors:

$$\text{WorkerFlows}_{lk} = \beta_1 \text{Distance}_{lk} + \beta_2 |\Delta\text{HousingPrice}_{lk}| + \beta_3 |\Delta\text{Amenities}_{lk}| + \beta_4 \text{TotalCalls}_{lk} + \mu_l + \mu_k + \epsilon_{lk}$$

The prediction exercise estimates the same first-half-sample specifications and predicts second-half worker flows (Table 2, Panel B, p. 1182). Standard errors in Panel A are two-way clustered by area $l$ and area $k$; there are 987,713 neighborhood-pair observations.

Main referral regression (eq. 1, p. 1183-1184). Let $\text{M}_{il} = 1$ if job switcher $i$ moves to location $l$ within the new workplace neighborhood. The specification restricts individual $i$‘s choice set to locations within the destination neighborhood to absorb heterogeneity across neighborhoods:

M_{il} = \beta \text{Friend}_{il} + \mathbf{X}_i \mathbf{Z}_l \gamma + \lambda_{\tilde{c},c} + \epsilon_{il} \tag{1}

where $\text{Friend}_{il} = 1$ if at least one of $i$‘s social contacts works at location $l$ three months before the job switch; $\mathbf{X}_i$ = individual demographics (gender, age groups, migration status, total social contacts); $\mathbf{Z}_l$ = location amenities (restaurants, roads and parking lots, schools within 500 m radius); $\lambda_{\tilde{c},c}$ = old-by-new neighborhood-pair fixed effects (20,811 total pairs in unrestricted sample; 16,468 in the baseline vacancy-restricted sample). Standard errors are clustered at the neighborhood-pair level.

Event study (p. 1187). Call frequency between switcher $i$ and friend $j$ in month $t$ is regressed on event-time dummies interacted with referrer vs non-referrer status, covering an event window from 11 months before to 9 months after the job switch (month $s = -1$ is the reference category):

$\text{Freq}_{ijt} = \sum_{s=-11}^{9} \gamma_s \text{Referral}_{ij} \cdot \mathbf{1}\{t = s\} + \sum_{\substack{s=-11 \\ s \neq -1}}^{9} b_s \text{Nonreferral}_{ij} \cdot \mathbf{1}\{t = s\} + \lambda_i + \tau_t + \epsilon_{ijt}$

where $\lambda_i$ are individual fixed effects and $\tau_t$ are calendar month fixed effects. The coefficients $\{\gamma_s, b_s\}$ capture changes in call frequency relative to the individual’s own baseline rate of talking to non-referrer friends. Standard errors are clustered at the individual level.

Worker outcome regression (eq. 2, p. 1197). Labor market outcomes of the referral job:

Y_{ilr} = \beta \text{Friend}_{ilr} + \mathbf{X}_i \mathbf{Z}_l \gamma + \lambda_c + \alpha_r + \epsilon_{ilr} \tag{2}

where $Y_{ilr}$ is a labor outcome for worker $i$ at new work location $l$ in residential neighborhood $r$; $\lambda_c$ = new work neighborhood fixed effect; $\alpha_r$ = residential neighborhood fixed effect; controls $\mathbf{X}_i$ include gender, age groups, migration status, and log number of social contacts. Standard errors are two-way clustered by residential and new work neighborhood.

Firm performance regression (eq. 3, p. 1200). Log firm outcomes:

Y_i = \gamma \text{Referral}_i + \mathbf{Z}_i \beta + \lambda_c + \epsilon_i \tag{3}

where $\text{Referral}_i = 1$ if at least one new hire at firm $i$ has a social contact already working there; $Y_i \in \{\log(\text{net inflow}),\, \log(\text{matching rate}),\, \log(\text{growth rate})\}$; $\lambda_c$ = neighborhood fixed effect; $\mathbf{Z}_i$ includes firm age, 18 industry dummies, SOE dummy, average employees 2010-2015, average capital stock 2010-2015, previous employment growth rate, share of female workers, share of migrants, average employee age, average employee housing price, and the firm’s referral network size. Standard errors are clustered at the neighborhood level.

Empirical specifications

Baseline specification (Table 3, p. 1186). Equation (1) estimated on 915,251 switcher-location pairs (restricted to individuals facing at least one alternative same-neighborhood, same-occupation, same-salary-range opening). Old-by-new neighborhood-pair fixed effects (16,468 pairs). The mean within-neighborhood switching probability is 0.09. The baseline referral coefficient is 0.35 (col 2). Columns 3-4 add falsification friend types (moved-away: 0.07; lives-but-not-works: 0.15); column 5 replaces direct friends with friends-of-friends (coefficient 0.14, confirming these second-degree links carry less job information); column 6 adds extensive local labor market controls (0.33, similar to baseline).

Effect heterogeneity (Table 4, p. 1191). Equation (1) augmented with interaction terms $\text{Friend}_{il} \times X_i$, one per column: (i) distance between old and new workplaces, coefficient 0.002 (SE 0.0004); (ii) distance between home and new workplace, 0.002 (SE 0.0003); (iii) young dummy (ages 25-34), +0.04 (SE 0.01); (iv) rural-to-urban switch, +0.32 (SE 0.05); (v) sector-change dummy, +0.21 (SE 0.02). For rural-to-urban movers and sector-changers the point estimates of the total referral effect are 0.66 and 0.53 respectively, substantially above the base estimate.

Event study (Figure 3, p. 1188). 238,092 switcher-referrer-month observations and 4,759,176 switcher-nonreferrer-month observations. The referrer-pair call frequency coefficient rises from near zero in the pre-event months to a peak of approximately +8 above baseline just before the job switch, then remains elevated post-switch as referrers become coworkers. Nonreferrer friends show coefficients near zero throughout. Falsification event studies (Figure 4, p. 1190) show that moved-away and lives-at-new-location friends display flat or mildly elevated patterns with no inverted U-shape, confirming the information spike is specific to current employment at the destination.

Comparison with literature proxies (Table 5, p. 1193). Equation (1) replaces the call-based Friend dummy with (i) residential neighbor dummy (0.21, SE 0.01) and (ii) same-birth-county coworker dummy (0.10, SE 0.01). Columns 3 and 4 each include one proxy alongside the call-based measure. In column 3, the residential neighbor coefficient falls to 0.18 (SE 0.01) while the call-based Friend (not neighbor) coefficient is 0.25 (SE 0.01). In column 4, the same-birth-county coefficient falls to 0.09 (SE 0.01) while the call-based Friend (not same birth county) coefficient is 0.35 (SE 0.03). Both proxy estimates decline when the direct communication measure is included, indicating that the proxy measures are weaker predictors in this sample.

Placebo friend types (Table 3, p. 1186; Figure 4, pp. 1189-1190). Table 3 adds one placebo indicator to the current-referrer indicator in each of columns 3-5. Written out from those separate Table 3 specifications:

$$M_{il} = \beta_F \text{Friend}_{il} + \beta_P \text{PlaceboFriendType}_{il} + \mathbf{X}_i \mathbf{Z}_l \gamma + \lambda_{\tilde{c},c} + \epsilon_{il}$$

The placebo indicator denotes a moved-away friend in column 3, a friend living but not working at the new location in column 4, or a friend of a nonreferrer friend in column 5.

Figure 4 reports two unnumbered event-study specifications (pp. 1188-1189), each with the same event-time range and fixed effects as Figure 3, one for moved-away friends and one for friends living at the new-workplace location without working there:

$$\text{Freq}_{ijt} = \sum_{s=-11}^{9} \gamma_s \text{Referral}_{ij} \cdot \mathbf{1}\{t=s\} + \sum_{s=-11}^{9} \alpha_s \text{MovedAway}_{ij} \cdot \mathbf{1}\{t=s\} + \sum_{\substack{s=-11 \\ s \neq -1}}^{9} b_s \text{OtherFriends}_{ij} \cdot \mathbf{1}\{t=s\} + \lambda_i + \tau_t + \epsilon_{ijt}$$ $$\text{Freq}_{ijt} = \sum_{s=-11}^{9} \gamma_s \text{Referral}_{ij} \cdot \mathbf{1}\{t=s\} + \sum_{s=-11}^{9} \alpha_s \text{LiveAtNewPlace}_{ij} \cdot \mathbf{1}\{t=s\} + \sum_{\substack{s=-11 \\ s \neq -1}}^{9} b_s \text{OtherFriends}_{ij} \cdot \mathbf{1}\{t=s\} + \lambda_i + \tau_t + \epsilon_{ijt}$$

Event-study errors are clustered by individual; Table 3 errors are clustered by neighborhood pair. The location-choice placebo regressions use the 915,251 switcher-location pairs and the old-by-new neighborhood-pair fixed effects.

Information-asymmetry heterogeneity (Table 4, p. 1191). Each column augments equation (1) with one interaction, with the interaction variable being old-to-new workplace distance, home-to-new workplace distance, young (ages 25-34), rural-to-urban move, or sector change:

$$M_{il} = \beta \text{Friend}_{il} + \delta (\text{Friend}_{il} \times H_i) + \mathbf{X}_i \mathbf{Z}_l \gamma + \lambda_{\tilde{c},c} + \epsilon_{il}$$

The Table 4 sample is 915,251 switcher-location pairs with 16,468 old-by-new neighborhood-pair fixed effects; standard errors are clustered by neighborhood pair. Columns 2-6 also control for the baseline level of the interacted variable.

Proxy comparisons (Table 5, p. 1193). Columns 1 and 3 use residential-neighbor indicators, while columns 2 and 4 use same-birth-county coworker indicators. The joint columns separate the proxy tie from direct communication ties:

$$M_{il} = \beta_1 \text{Neighbor}_{il} + \beta_2 \text{FriendNotNeighbor}_{il} + \mathbf{X}_i \mathbf{Z}_l \gamma + \lambda_{\tilde{c},c} + \alpha_r + \epsilon_{il}$$ $$M_{il} = \beta_1 \text{SameBirthCounty}_{il} + \beta_2 \text{FriendNotSameBirthCounty}_{il} + \mathbf{X}_i \mathbf{Z}_l \gamma + \lambda_{\tilde{c},c} + \alpha_b + \epsilon_{il}$$

Each uses 915,251 observations and old-by-new work-neighborhood fixed effects. Columns 1 and 3 include residential-neighborhood fixed effects; columns 2 and 4 include birth-county fixed effects. Errors cluster by old-by-new work-neighborhood pair and residential neighborhood or birth county, respectively.

Homophily and preference controls (Table 6, p. 1195). These are the baseline location-choice specification with additional controls for observed similarity, 100 switcher-friend clusters, or indicators for having more friends and more similar friends at the new workplace than at the old workplace. In column 3, the reported more-friends indicator coefficient is 0.002 (SE 0.001). Each column has 915,251 switcher-location pairs, 16,468 old-by-new neighborhood-pair fixed effects, and standard errors clustered by neighborhood pair. Written out from the Table 6 control sets (the table does not number this specification):

$$M_{il} = \beta \text{Friend}_{il} + \mathbf{H}_{il}'\theta + \mathbf{X}_i \mathbf{Z}_l \gamma + \lambda_{\tilde{c},c} + \epsilon_{il}$$

Here, $\mathbf{H}_{il}$ contains observed homophily measures, same-cluster indicators, or the indicators for more friends and more similar friends at the new workplace, depending on the column.

Tie strength and contact diversity (text pp. 1193-1194; online Appendix Tables S9-S10). The paper contrasts Granovetter (1973)‘s argument that weak ties can provide diverse information with Gee et al. (2017)‘s evidence that stronger ties matter more at the margin; it finds larger referral effects for stronger ties. The tie-strength test interacts the friend indicator with call intensity; the diversity test relates successful referral job changes to normalized social and income entropy, total calls, strong-tie share, and demographics. These descriptions are written out from the text and are not numbered equations:

$$M_{il} = \beta \text{Friend}_{il} + \delta (\text{Friend}_{il} \times \text{TieStrength}_{il}) + \mathbf{X}_i \mathbf{Z}_l \gamma + \lambda_{\tilde{c},c} + \epsilon_{il}$$ $$M_i = \beta_1 \text{SocialEntropy}_i + \beta_2 \text{IncomeEntropy}_i + \beta_3 \text{TotalCalls}_i + \beta_4 \text{StrongTieShare}_i + \mathbf{X}_i'\gamma + \epsilon_i$$

Worker benefits (Table 7, p. 1199). Equation (2) estimated separately for five outcome variables. Sample sizes vary by outcome due to data availability. All columns include residential and new-work neighborhood fixed effects. Key results: wage = 0.62 thousand RMB (SE 0.31, N=17,615); coworker housing price difference = 0.07 thousand RMB/per square meter (SE 0.04, N=23,323); PT-to-FT = 0.014 (SE 0.007, N=19,431); shorter commute = 0.09 (SE 0.01, N=29,117); non-SOE to SOE = 0.012 (SE 0.005, N=15,881).

Firm benefits (Table 8, p. 1201). Equation (3) estimated on large-firm locations (firms with more than 100 employees) to reduce spurious worker-firm linking. In the most saturated specification (column 4/8/12 for each panel), gamma = 0.63 (SE 0.14) for log net inflow, 0.84 (SE 0.27) for log matching rate, and 0.45 (SE 0.11) for log firm growth rate. The estimates are stable across specifications with progressively richer firm and employee controls, arguing against upward bias from fast-growing firms being more likely to use referrals.

Unemployment and reemployment event study (Figure 5, p. 1204). The paper reports this unnumbered specification for 3,638 workers with an employment gap, with individual and calendar-month fixed effects. Event indices are relative to unemployment (s) and reemployment (r); the reference is nonreferrer calls in the month before unemployment:

$$\text{Freq}_{ijt} = \sum_{s=-5}^{3} \gamma_s \text{Referral}_{ij} \cdot \mathbf{1}\{t=s, r<0\} + \sum_{r=-3}^{4} \gamma_r \text{Referral}_{ij} \cdot \mathbf{1}\{t=r, s\geq 0\} + \sum_{\substack{s=-5 \\ s\neq -1}}^{3} b_s \text{Nonreferral}_{ij} \cdot \mathbf{1}\{t=s, r<0\} + \sum_{r=-3}^{4} b_r \text{Nonreferral}_{ij} \cdot \mathbf{1}\{t=r, s\geq 0\} + \lambda_i + \tau_t + \epsilon_{ijt}$$

The event-study errors are clustered by individual; the figure reports 53,244 referrer-month and 747,804 nonreferrer-month observations.

Datasets used

Dataset	Role in paper	Wiki page
Geocoded cellphone records, Company A (anonymous northern Chinese city)	Main analysis: social network construction, information-flow measures, work and home location histories for 456,000 users, Nov 2016-Oct 2017	No page yet (proprietary; provider anonymous under data-sharing agreement)
Administrative firm-level records (merged by location)	Industry composition, average payroll, number of employees, capital stock; used for firm-performance regressions and location controls	No page yet (Chinese administrative data; not publicly available)
Job postings data (unnamed online platform)	Occupation and salary range at each location; used to restrict baseline sample to switchers facing comparable alternative opportunities	No page yet
Residential housing price data	Proxy for coworker socioeconomic status; used as a nonwage benefit measure (delta coworker HP)	No page yet
China Family Panel Studies (CFPS, 2014)	Descriptive only: national average demographics and job-search method frequencies for comparison with sample (Figure 1, Table 1)	No page yet
US Current Population Survey (2014)	Descriptive only: US job-search method frequencies for cross-country comparison (Figure 1)	No page yet
Communication-channel survey of 20,000 cellphone users in a comparable Chinese city (November 2020)	External validation of the relation between calls, text messages, and WeChat use (online Appendix Table S18)	No page yet

Sample: November 2016 to October 2017 (12 months). Final analysis sample: 456,000 individuals with stable work locations for at least 45 weeks and at most two work locations; 38,102 job switchers (8 percent of sample). Social contacts defined as anyone with at least one call to or from individual i in the three months prior to the job switch; on average 50 percent of a user’s friends are Company A customers.

When to read the full paper

Use the original if you are: measuring information transmission in labor markets with mobile phone data or other digital footprints; designing or evaluating employee referral programs and need evidence on the heterogeneity of referral effects by worker type; working on urban labor mobility in developing economies where formal job-search institutions are weak; extending the event-study design to other communication technologies (WeChat, messaging apps) or other information channels; or benchmarking referral-effect magnitudes for structural job-search models. The online appendix (referenced in the paper and in the replication package at 10.3886/E183161V1) contains detailed robustness tables and the event study for unemployed job-seekers.

Attribution and rights

Source: peer-reviewed, American Economic Review 113(5), May 2023. Replication data available at 10.3886/E183161V1. This distillation was extracted by an LLM on 2026-06-24 and is not human-verified or independently reproduced. No CC licence was found in Crossref metadata; standard AEA copyright applies; extract-only.

Barwick, Panle Jia, Yanyan Liu, Eleonora Patacchini, and Qi Wu. “Information, Mobile Communication, and Referral Effects.” American Economic Review 113, no. 5 (May 2023): 1170-1207. DOI: 10.1257/aer.20200187. Copyright 2023 American Economic Association. This page is an extract only; it is not a substitute for the original.