## The Economic Consequences of Social Unrest: Evidence from Initial Public Offerings

## Source details

**Canonical URL:** [The Economic Consequences of Social Unrest: Evidence from Initial Public Offerings](https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023232-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2023/english/wpiea2023232-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2023/english/wpiea2023232-print-pdf.pdf.json)

---

### Key findings on social unrest and IPO underpricing
- Greater social unrest is associated with lower first-day returns (underpricing), suggesting investor sentiment effects dominate uncertainty effects.
- When controlling for attitudes toward uncertainty, stronger aversion to uncertainty and ambiguity dampens the negative relation between social unrest and underpricing.
- 2SLS estimation using motivated instruments confirms the negative relation between social unrest and underpricing.
- The negative association between unrest and underpricing is weaker in places that score better on: control of corruption, government effectiveness, political stability, regulatory quality, rule of law, and voice and accountability.
- Short-selling constraints magnify the negative relation between social unrest and underpricing: short selling bans, security lending bans, and markets where short selling is not typically practiced aggravate the effect.
- Evidence indicates social unrest’s negative impact on investor sentiment tends to dominate any increase in uncertainty it might cause with respect to IPO underpricing.

### Data, measures, and sample
- Sample: 15,145 IPOs issued in 36 markets between 1998 and 2018.
- Data sources: Refinitiv SDC Platinum for offerings; Datastream for secondary market prices; SEDOL identifier used to match IPOs to Datastream.
- Underpricing measure: Underpricing = (First-day closing price – IPO offer price) / IPO offer price. For France, Greece, and Thailand, the tenth valid secondary market closing price is used due to daily volatility limits.
- RSUI (Reported Social Unrest Index):
  - Barrett et al.’s (2022a) RSUI provides monthly data for 130 economies from 1985 through 2022, constructed from English-language newspaper article searches for keywords such as “protest,” “riot,” “revolution,” and “unrest.”
  - RSUI rebased to average 100 for all sample markets across 1998-2018.
  - RSUI values in the sample range from zero to 4,055.35 (Netherlands, March 2017).
  - Market-level examples: Greece average RSUI value 14.90; Türkiye average RSUI value 252.79.
- Coverage notes: Nine markets in the sample experienced no unrest events during 1998-2018: Australia, Canada, Ireland, Japan, New Zealand, Norway, Singapore, South Korea, and the U.S.

### Empirical strategy and identification
- Baseline regression (location fixed effects):
  - Underpricing_ijt = β0 + β1 log(1 + RSUI_jt) + β X_ijt + μ_i + γ_j + α_t + ε_ijt
  - Underpricing_ijt is the first-day return for IPO of firm i in location j in month t.
  - RSUI_jt is log-transformed.
  - X_ijt includes firm-, deal-, and market-level covariates; μ_i, γ_j, and α_t are industry, location, and issue year effects.
- Instrumental variables (2SLS):
  - Neighbors’ incidents of social unrest in the preceding six months instrument for local social unrest.
  - Misery index (unemployment rate plus inflation rate) used as a second instrument.
  - 2SLS addresses endogeneity and confirms the negative RSUI–underpricing relation.

### Conceptual framework and mechanisms
- Stylized model highlights two opposing channels:
  - Lower realized investor sentiment (휖) reduces post-IPO returns.
  - Higher perceived uncertainty (휎2) increases IPO discounts and thus raises post-IPO returns.
- Post-IPO return expression:
  - r = log(p1/p0) = log(1 + 휖) − log(1 − γ휎2)
- Mechanistic implications:
  - Negative realized sentiment lowers returns; higher perceived uncertainty raises returns.
  - Timing distinction: uncertainty is ex ante (incorporated in p0), sentiment is ex post (realized in p1).
- Framework particularly well-suited for IPOs because seller likely does not know ex post sentiment at offering.

### Empirical results — magnitudes and robustness
- Descriptive comparisons:
  - IPOs issued within three months of a social unrest event: average underpricing 23.3%.
  - All other IPOs: average underpricing 31.4%.
  - Difference: approximately eight percentage points lower underpricing following unrest events.
  - Offer size: Inflation-adjusted average offer size = $112.62 million (similar across subsamples).
  - RSUI: Average RSUI value is over seven times higher for IPOs issued in the three months following an unrest event.
  - Market-specific first-day returns: Mainland China average first-day return = 62.00%; Japan average first-day return = 72.21%.
- Main regression (Table 5) — RSUI and underpricing:
  - RSUI coefficient examples: -0.0089** (Location FE), -0.0083** (Location FE alternative), -0.0052** (HLM), -0.0149*** (WLS).
  - Offer size (log) coefficient: -0.0437***.
  - Democracy coefficient examples: 0.3936***, 0.3900***, 0.2646**.
  - Observations: 15,145; example R-squared: 0.137, 0.144.
- Instrumental-variable (2SLS) results (Table 6):
  - First-stage: neighbor unrest and neighbor unrest (#) positive and significant; misery index positively associated with social unrest.
  - First-stage RSUI’ coefficients: -0.0867* and -0.1215*** (reported).
  - Anderson χ2: 73.64 and 119.149.
  - Cragg-Donald F-stat: 36.89 and 59.87; Stock-Yogo (10%) critical value 19.93.
  - Second-stage: RSUI′ coefficient negative and larger in magnitude than OLS.
  - Observations: 15,143; First-stage/Second-stage R-squared reported: 0.109 and 0.078.
- Institutional quality (Table 7):
  - RSUI main effects remain negative and significant across measures (examples): Control of corruption RSUI -0.0345***; Government effectiveness RSUI -0.0556***; Political stability RSUI -0.0260***.
  - Interaction terms (RSUI × high-quality institution) are positive and significant, indicating higher-quality institutions weaken the negative RSUI–underpricing relation.
  - Observations: 15,145; Number of locations: 36.
- Limits to arbitrage (Table 8):
  - Short-selling constraint prevalence: 18.4% short selling ban; 20.3% security lending ban; 34.4% short selling not typically practiced.
  - RSUI × Short selling measure coefficients: -0.0493*** (Short selling banned), -0.0485*** (Security lending banned), -0.0316*** (Short selling not practiced).
  - Interaction terms negative and highly significant — short-sale constraints magnify the negative RSUI–underpricing relation.
  - Observations: 15,145; Number of locations: 36.
- Nonlinearity (Table 9):
  - Regression including RSUI and RSUI^2 yields inverse U-shaped relation.
  - RSUI coefficient: 0.0156 (t-stat 1.54).
  - RSUI2 coefficient: -0.0040*** (t-stat -2.58).
  - Predicted pattern: at low RSUI, uncertainty effect can dominate (increasing underpricing); at (log) RSUI values > 3.10, investor sentiment effect dominates (lowering underpricing).
  - Observations: 15,145; R-squared 0.137.
- Additional robustness (Table 10 and Table 11):
  - Alternative dependent variables:
    - Two-week return RSUI -0.0132** (t-stat -2.26).
    - One-month return RSUI -0.0102* (t-stat -1.88).
    - Offer-to-open return RSUI -0.0108*** (t-stat -3.40).
    - Open-to-close return RSUI -0.0017 (t-stat -1.42).
  - Influence of specific markets:
    - Excluding Mainland China: RSUI -0.0060* (t-stat -1.73).
    - Excluding Hong Kong SAR: RSUI -0.0062* (t-stat -1.70).
    - Excluding U.S.: RSUI -0.0088** (t-stat -2.42).
  - Excluding nine markets with no reported incidents of unrest — negative relation persists.
  - Conclusion: Negative RSUI–underpricing relation robust to alternative return measures and exclusions.

### Contributions, interpretation, and practical implications
- Contributions:
  - Documents novel evidence that social unrest affects IPO underpricing via investor sentiment channels, with the investor-sentiment reduction typically outweighing uncertainty-induced underpricing increases.
  - Links social unrest to ESG considerations and potential capital flow consequences via ESG investment channels.
  - Identifies institutional quality and market frictions (short-selling constraints) as important moderators.
- Practical implications:
  - IPO issuers and advisors should weigh competing effects of investor sentiment and uncertainty when timing and pricing offerings amid social unrest.
  - Investors should consider how social unrest may reduce investor sentiment and thereby lower expected first-day returns.

*Source: IMF Working Paper — The Economic Consequences of Social Unrest: Evidence from Initial Public Offerings (excerpt provided).*

### 2018. Using Barrett et al.’s (2022a) RSUI to measure social unrest in the month and location of

### The Economic Consequences of Social Unrest: Evidence from Initial Public Offerings

### Key findings on social unrest and IPO underpricing
- Greater social unrest is associated with lower first-day returns (underpricing), suggesting investor sentiment effects dominate uncertainty effects.
- When controlling for attitudes toward uncertainty, stronger aversion to uncertainty and ambiguity dampens the negative relation between social unrest and underpricing.
- 2SLS estimation using motivated instruments confirms the negative relation between social unrest and underpricing.
- The negative association between unrest and underpricing is weaker in places that score better on: control of corruption, government effectiveness, political stability, regulatory quality, rule of law, and voice and accountability.
- Short-selling constraints magnify the negative relation between social unrest and underpricing: short selling bans, security lending bans, and markets where short selling is not typically practiced aggravate the effect.
- Evidence indicates social unrest’s negative impact on investor sentiment tends to dominate any increase in uncertainty it might cause with respect to IPO underpricing.

### Data, measures, and sample
- Sample: 15,145 IPOs issued in 36 markets between 1998 and 2018.
- Data sources: Refinitiv SDC Platinum for offerings; Datastream for secondary market prices; SEDOL identifier used to match IPOs to Datastream.
- Underpricing measure: Underpricing = (First-day closing price – IPO offer price) / IPO offer price. For France, Greece, and Thailand, the tenth valid secondary market closing price is used due to daily volatility limits.
- RSUI (Reported Social Unrest Index):
  - Used Barrett et al.’s (2022a) RSUI as the primary measure of social unrest.
  - RSUI provides monthly data for 130 economies from 1985 through 2022, constructed from English-language newspaper article searches for keywords such as “protest,” “riot,” “revolution,” and “unrest.”
  - RSUI rebased to average 100 for all sample markets across 1998-2018.
  - RSUI values in the sample range from zero to 4,055.35 (Netherlands, March 2017).
  - Market-level examples: Greece average RSUI value 14.90; Türkiye average RSUI value 252.79.
- Coverage notes: Nine markets in the sample experienced no unrest events during 1998-2018: Australia, Canada, Ireland, Japan, New Zealand, Norway, Singapore, South Korea, and the U.S.

### Empirical strategy and identification
- Baseline regression specification (location fixed effects):
  - Underpricing_ijt = β0 + β1 log(1 + RSUI_jt) + β X_ijt + μ_i + γ_j + α_t + ε_ijt
  - Underpricing_ijt is the first-day return for IPO of firm i in location j in month t.
  - RSUI_jt is the Reported Social Unrest Index for location j in month t; log transformation used.
  - X_ijt is a vector of covariates; μ_i, γ_j, and α_t are industry, location, and issue year effects.
- Instrumental variables (2SLS):
  - Neighbors’ incidents of social unrest in the preceding six months instrument for local social unrest (motivated by Barrett et al. (2022a) finding that unrest increases potential for unrest in neighboring countries over the following six months by one percentage point).
  - Misery index (unemployment rate plus inflation rate) for the IPO firm’s economy used as a second instrument (motivated by Redl and Hlatshwayo (2021) that prior economic conditions foment unrest).
  - 2SLS results confirm the negative relation between social unrest and underpricing, addressing model misspecification concerns.

### Institutional quality, market features, and mechanisms
- Institutional quality moderates the effect:
  - Negative association between unrest and underpricing is weaker in markets with higher scores on control of corruption, government effectiveness, political stability, regulatory quality, rule of law, and voice and accountability.
  - Interpreted as high-quality institutions weakening the association between investor sentiment and IPO returns.
- Short selling and arbitrage constraints:
  - Consistent with Baker and Wurgler (2006) and Boulton et al. (2020), short-sale constraints exacerbate the impact of investor sentiment on IPO mispricing.
  - Depending on the measure, short selling constraints affect 18-34% of sample IPOs.
  - Short sale bans, security lending bans, and markets where short selling is not typically practiced aggravate the negative relation between social unrest and underpricing.

### Descriptive statistics and sample comparisons (IPOs near unrest vs others)
- Underpricing:
  - IPOs issued within three months of a social unrest event: average underpricing 23.3%.
  - All other IPOs: average underpricing 31.4%.
  - Difference: approximately eight percentage points lower underpricing following unrest events.
- RSUI:
  - Average RSUI value is over seven times higher for IPOs issued in the three months following an unrest event.
- Offer size:
  - Inflation-adjusted average offer size = $112.62 million (similar across subsamples).
- Intermediaries and deal characteristics (following unrest vs quieter times):
  - Top underwriter employed: 12.5% versus 24.3%.
  - VC backed: 16.8% versus 22.2%.
  - Lockup length: average approximately 17.6 days longer for IPOs issued following an unrest event.
  - Bookbuilt IPOs: 47.8% versus 65.4%.
  - Firm commitment offerings: 59.8% versus 65.7%.
  - Carve-outs: 20.1% versus 9.8%.
  - High tech IPOs: 12.5% versus 20.3%.
- Market-specific first-day returns:
  - Mainland China average first-day return = 62.00%.
  - Japan average first-day return = 72.21%.

### Contributions and implications
- Provides novel evidence that social unrest affects young firms’ ability to raise capital via IPO underpricing through investor sentiment channels.
- Links social unrest to ESG considerations and possible capital flow consequences via ESG investment channels.
- Sheds light on how institutional quality and market frictions (short-selling constraints) shape the financial consequences of social unrest for IPOs.

*Source: IMF Working Paper — The Economic Consequences of Social Unrest: Evidence from Initial Public Offerings (excerpt provided).*

### 0.153  for  the  remaining  variables. The  low  correlations,  coupled  with an  average (maximum)

### 3. A stylized model of post-IPO returns

### Key empirical diagnostic
- Variance inflation factor: average (maximum) variance inflation factor of 3.2 (6.38).
- Correlations: 0.153 for the remaining variables.
- Interpretation: The low correlations, coupled with an average (maximum) variance inflation factor of 3.2 (6.38), suggest that multicollinearity is not a significant concern in our setting.

### Conceptual framework: channels affecting post-IPO returns
- Two main forces during times of social unrest:
  - Lower investor sentiment (휖) — an ex post phenomenon (realization of a shock) that tends to lower post-IPO returns when worse.
  - Higher uncertainty (휎2) — an ex ante phenomenon (perceived variance) that tends to raise post-IPO returns because IPO price incorporates a discount for uncertainty.
- Timing distinction:
  - Uncertainty is a property of the perceived distribution of risks before realization.
  - Sentiment is realized after the IPO; if known at IPO time, it would be incorporated into the IPO price and thus not affect returns.

### Model setup and expressions
- Post-IPO price relation:
  - 푝1 = 푝1^e (1 + 휖)
  - 휖 is a random variable capturing realized investor sentiment.
- IPO price set with a discount reflecting perceived uncertainty:
  - 푝0 = 푝1^e (1 − 훾휎2)
  - 훾 reflects the extent to which purchasers in the IPO must be compensated for bearing risk.
  - 휎2 is the IPO seller’s perceived variance of the post-IPO price.
  - If sellers were fully informed and rational then 휎2 = Var⁡휖, although the model does not impose this equality.
- Post-IPO return (extent of underpricing):
  - 푟 = log(푝1/푝0) = log(1 + 휖) − log(1 − 훾휎2)

### Mechanistic implications (as derived from the model)
- Post-IPO returns move with realized sentiment and perceived uncertainty:
  - Returns are lower when investor sentiment 휖 is worse (negative realizations of 휖 reduce log(1 + 휖)).
  - Returns are higher when uncertainty 휎2 is high (larger 훾휎2 increases the IPO discount, raising realized returns conditional on post-IPO price).
- Changes in expected future price 푝1^e have no effect on returns because they proportionally affect both 푝0 and 푝1 and cancel out in the return expression.
- The multiplicative specification for sentiment is chosen for algebraic convenience; an additive representation would be equivalent if the variance of sentiment were scaled by 푝1^e.
- Linearity and microfoundations are not essential to the illustrative example; the chosen forms make the countervailing mechanisms stark and clear.

### Suitability for IPOs versus other assets
- The framework is particularly well-suited to IPO returns because:
  - Investor sentiments can reasonably be taken as unknown to the seller at the time of the IPO, allowing sentiment shocks to affect returns.
  - For other assets with trading histories, sentiment is more likely to be well-understood and incorporated into prices (analogous to changes in 푝1^e), thus having less impact on returns.

*IMF WORKING PAPERS — The Economic Consequences of Social Unrest: Evidence from Initial Public Offerings*

### 4. Empirical results

### 4. Empirical results

### 4.1. IPO characteristics following unrest events
- Logistic regressions (odds ratios) compare IPOs issued within three and six months of a social unrest incident to IPOs issued during more stable periods; models include location fixed (random) effects.
- Findings:
  - Stronger IPO activity and higher market returns decrease the likelihood that an IPO is issued in the months following unrest.
  - IPOs issued after incidents of unrest are more common in more liquid markets (odds ratio for liquidity > 1).
  - Bookbuilt offerings: odds ratios are all less than one, indicating bookbuilt offerings are approximately 50 percent less common following incidents of unrest.
  - Some evidence (in some models) that IPO size, VC backing, firm commitment offerings, and carve-outs are associated with the likelihood an IPO follows unrest.
- Overall: Except for weaker prevailing market conditions (fewer IPOs and lower market returns) and the lower incidence of bookbuilt offerings, IPOs issued following periods of unrest do not appear markedly different than IPOs issued at other times.

### 4.2. Social unrest and underpricing
- Competing hypotheses:
  - Uncertainty channel: social unrest increases uncertainty → more investor information gathering → increases underpricing (Beatty and Ritter, 1986; Ritter, 1984).
  - Investor sentiment channel: social unrest lowers investor sentiment and aftermarket prices → reduces underpricing (Derrien, 2005; Cornelli et al., 2006).
- Multivariate results (Table 5, baseline location fixed-effects model, Eq. (2)):
  - Coefficient for RSUI is negative and significant (p-value = 0.012), suggesting the investor sentiment effect dominates.
  - Contemporaneous unrest drives the negative relation: residual from regressing contemporaneous RSUI on lagged RSUI is negative and significant; lagged RSUI is not statistically significant.
  - Results robust to hierarchical linear modeling (HLM) and weighted least squares (WLS): RSUI coefficient remains negative and significant.
  - Interaction with above-median uncertainty avoidance (Hofstede, 2001): RSUI coefficient negative; interaction term positive, indicating the RSUI effect on underpricing is weaker where uncertainty avoidance is high.
- Control-variable patterns:
  - IPOs are underpriced less in economies with higher financial integration and economic freedom scores.
  - IPOs are underpriced more in economies with higher democracy scores.
  - IPO activity and market return exhibit a hot market effect (positive coefficients).
  - Offer size: larger IPOs tend to be underpriced less.
  - Top underwriter and VC backed coefficients are positive (consistent with prior literature linking reputation/VC and underpricing).
  - Lockup length and underpricing are negatively correlated.
  - Bookbuilt offerings exhibit lower underpricing.
  - High tech firms are underpriced more than other industries.
- Conclusion: Negative association between unrest and underpricing indicates investor sentiment effects (lower first-day returns) dominate uncertainty effects.

### 4.3. Instrumental variable analysis
- Motivation: address omitted-variable bias and endogeneity concerns via 2SLS.
- Instruments:
  - Neighbor unrest: indicator = 1 if a neighbor experienced an unrest event in the preceding six months (alternative: neighbor unrest (#) = number of neighbors with unrest in preceding six months). Prior research: Barrett et al. (2022a) find unrest increases the potential for unrest in neighboring economies over the following six months by one percentage point.
  - Misery index: sum of unemployment rate and inflation rate.
- Two-stage model (location fixed effects):
  - First stage: RSUI regressed on neighbor unrest, misery index, controls, and fixed effects.
  - Second stage: Underpricing regressed on instrumented RSUI (RSUI′), controls, and fixed effects; log transformation of RSUI used.
- Results (Table 6):
  - First-stage: neighbor unrest and neighbor unrest (#) positive and significant; misery index positively associated with social unrest.
  - Second-stage: RSUI′ coefficient is negative, reinforcing the negative relation between social unrest and underpricing.
  - 2SLS yields larger RSUI coefficients than OLS (possible explanations: omitted-variable downward bias in OLS or contemporaneous RSUI measurement error biasing OLS toward zero).
  - Instrument validity tests: Significant Anderson and Hausman tests reject underidentification and confirm RSUI endogeneity; Anderson-Rubin, Cragg-Donald, and Stock-Yogo statistics reject weak instruments.

### 4.4. Institutional quality
- Hypotheses:
  - High-quality institutions may weaken the link between investor sentiment and IPO returns (per Schmeling, 2009), moderating the negative RSUI-underpricing relation.
  - Alternatively, high-quality institutions may reduce uncertainty, making investor sentiment effects more apparent.
- Institutional measures: Worldwide Governance Indicators — control of corruption, government effectiveness, political stability, regulatory quality, rule of law, voice and accountability (correlations with RSUI range from -0.193 for political stability to -0.118 for rule of law).
- Method: HLM with each institutional measure and its interaction with RSUI added separately (Table 7).
- Findings:
  - RSUI remains negative and significant in every model.
  - Interaction terms are positive across measures, indicating higher-quality institutions moderate (weaken) the negative relation between RSUI and underpricing.
  - Interpretation: high-quality institutions reduce the sensitivity of IPO returns to investor sentiment shocks.

### 4.5. Limits to arbitrage
- Theory: Short-selling constraints magnify the impact of investor sentiment on mispricing (Baker and Wurgler, 2006).
- Short-selling constraint indicators (Boulton et al., 2020 data):
  - 18.4% of IPOs in sample issued in markets with a short selling ban.
  - 20.3% of IPOs issued in markets with a security lending ban.
  - 34.4% of IPOs issued in markets where short selling is not typically practiced.
- Empirical approach: interact RSUI with short-sale constraint indicators (Table 8).
- Results:
  - RSUI coefficients not statistically significant alone in these models.
  - Interaction terms between RSUI and short-sale constraints are negative and highly significant, indicating short-sale constraints magnify the negative relation between social unrest and underpricing.
  - Linear combinations reported show that markets where short selling is constrained (short selling bans, security lending bans, or absence of short selling) drive the negative impact of social unrest on underpricing.

### 4.6. Nonlinearity
- Rationale: social unrest affects uncertainty and investor sentiment differently across intensity levels, implying possible nonlinear RSUI-underpricing relation.
- Evidence:
  - Predicted underpricing from regression including RSUI and RSUI^2 shows an inverse U-shaped relation (Figure 3).
  - At low RSUI, social unrest slightly increases underpricing (uncertainty effect dominates).
  - At (log) RSUI values greater than 3.10, greater unrest associated with lower underpricing (investor sentiment effect dominates).
- Full regression (Table 9) with RSUI and RSUI squared:
  - Coefficient on RSUI positive; coefficient on RSUI squared negative — consistent with inverse U-shape: uncertainty dominates at low unrest, investor sentiment dominates at high unrest.

### 4.7. Additional robustness
- Alternative underpricing measures (Table 10):
  - Use 10th and 22nd secondary market closing prices to measure underpricing beyond IPO date; results continue to support negative RSUI-underpricing relation, with slightly larger coefficient magnitudes than Table 5.
  - Decomposition of first-day return:
    - Offer-to-open return (offer price to first-day opening price) and open-to-close return (first-day opening to first-day closing).
    - Offer-to-open returns drive the negative RSUI-underpricing relation; no evidence that open-to-close (information cascades) contribute.
- Influence of specific markets (Table 11):
  - Excluding individually: Mainland China, Hong Kong SAR, U.K., U.S. (markets with large IPO counts) — negative RSUI-underpricing relation persists.
  - Excluding individually: Austria, Netherlands, Sweden, Türkiye (markets with notable average unrest) — negative relation persists.
  - Excluding nine markets with no reported incidents of unrest (Australia, Canada, Ireland, Japan, New Zealand, Norway, Singapore, South Korea, U.S.) — negative relation persists.
- Conclusion: Negative relation between social unrest and IPO underpricing is robust to alternative return measures and exclusion of potentially influential markets.

*Source: wpiea2023232-print-pdf - 4. Empirical results*

### 5. Conclusion

### 5. Conclusion

### Summary of key findings
- Using a large sample of IPOs issued in 36 markets between 1998 and 2018, the study finds robust evidence that greater social unrest is associated with lower IPO underpricing.
- The negative impact of social unrest on investor sentiment outweighs the effect of social unrest on uncertainty, leading to lower first-day returns for IPOs during periods of unrest.
- This is documented as a significant relation between social unrest and underpricing; to the authors’ knowledge, this is the first study to report this finding.

### Mechanisms and interpretation
- Investor sentiment channel:
  - Higher investor sentiment is expected to increase first-day returns (underpricing) because an increase during the first trading day raises the closing price on day one (Ljungqvist et al. (2006)).
  - Social unrest lowers investor sentiment; therefore, IPOs issued during unrest should exhibit lower first-day returns.
- Uncertainty channel:
  - Increased uncertainty raises incentives to gather information, which exacerbates information disparities among IPO participants and necessitates greater underpricing.
  - Theory predicts a positive association between social unrest and first-day returns if the uncertainty channel dominates.
- Empirical conclusion:
  - The empirical evidence indicates the investor-sentiment reduction dominates the uncertainty amplification, producing a net negative relation between social unrest and IPO underpricing.

### Robustness and identification
- The negative relation between social unrest and underpricing is confirmed in a 2SLS specification that instruments for social unrest using:
  - the social unrest of neighbors, and
  - the misery index,
  - with issue month included to address model misspecification concerns.
- Social unrest and underpricing remain negatively correlated in this instrumental-variable analysis.

### Heterogeneity and moderating factors
- Cultural/behavioral trait:
  - The negative association between unrest and underpricing is weaker in markets with greater uncertainty avoidance—presumably because social unrest more strongly amplifies uncertainty in these locations.
- Institutional quality:
  - The negative relation between social unrest and underpricing is weaker for IPOs issued in places with higher-quality institutions that weaken the link between investor sentiment and stock returns (Schmeling, 2009).
- Market frictions:
  - Limits to arbitrage amplify the negative relation between social unrest and underpricing, consistent with prior research.

### Implications for practice
- IPO issuers and advisors:
  - Must weigh the competing effects of investor sentiment and uncertainty when deciding whether and at what price to take firms public amid social unrest.
- Investors:
  - Must consider how social unrest may reduce investor sentiment and thereby lower expected first-day returns when deciding whether and at what price to invest in IPOs.

*Source: 5. Conclusion — The Economic Consequences of Social Unrest: Evidence from Initial Public Offerings (IMF Working Paper).*

### References

### References

### Bibliography (selected entries from the References section)
- Autore, D., Boulton, T., Smart, S., & Zutter, C. (2014). The impact of institutional quality on initial public offerings. Journal of Economic and Business, 73, 65–96.
- Bajo, E., & Raimondo, C. (2017). Media sentiment and IPO underpricing. Journal of Corporate Finance, 46, 139‒153.
- Baker, E., Boulton, T., Braga-Alves, M., & Morey, M. (2021). ESG government risk and international IPO underpricing. Journal of Corporate Finance, 67, 101913.
- Baker, M., & Wurgler, J. (2006). Investor sentiment and the cross-section of stock returns. Journal of Finance, 61, 1645‒1680.
- Baker, M., & Wurgler, J. (2007). Investor sentiment in the stock market. Journal of Economic Perspectives, 21, 129‒151.
- Barrett, P., Appendino, M., Nguyen, K., & Miranda, J. (2022a). Measuring social unrest using media reports. Journal of Development Economics, 158, 102924.
- Barrett, P., Bodnar, M., Chen, S., & Chivakul, M. (2022b). Pricing protest: The response of financial markets to social unrest. IMF Working Paper.
- Barry, C., & Jennings, R. (1993). The opening price performance of initial public offerings of common stock. Financial Management, 22, 54–63.
- Barry, C., Muscarella, C., Peavy, J., & Vetsuypens, M. (1990). The role of venture capital in the creation of public companies. Journal of Financial Economics, 27, 447–471.
- Beatty, R., & Ritter, J. (1986). Investment banking, reputation, and the underpricing of initial public offerings. Journal of Financial Economics, 15, 213–232.
- Beatty, R., & Welch, I. (1996). Issuer expenses and legal liability in initial public offerings. Journal of Law and Economics, 39, 545–602.
- Benveniste, L. & Spindt, P. (1989). How investment bankers determine the offer price and allocate new issues. Journal of Financial Economics, 24, 343–361.
- Bhattacharya, U., Daouk, H., & Welker, M. (2003). The world price of earnings opacity. The Accounting Review, 78, 641‒678.
- Booth, J., & Smith, R. (1986). Capital raising, underwriting and the certification hypothesis. Journal of Financial Economics, 15, 261‒281.
- Borisov, A., Ellul, A., & Sevilir, M. (2021). Access to public capital markets and employment growth. Journal of Financial Economics, 141, 896‒918.
- Boulton, T. (2022). Economic policy uncertainty and international IPO underpricing. Journal of International Financial Markets, Institutions & Money, 81, 101689.
- Boulton, T., Smart, S., & Zutter, C. (2010). IPO underpricing and international corporate governance. Journal of International Business Studies, 41, 206–222.
- Boulton, T., Smart, S., & Zutter, C. (2011). Earnings quality and international IPO underpricing. The Accounting Review, 86, 483–505.
- Boulton, T., Smart, S., & Zutter, C. (2017). Conservatism and international IPO underpricing. Journal of International Business Studies, 48, 763–785.
- Boulton, T., Smart, S., & Zutter, C. (2020). Worldwide short selling regulations and IPO underpricing. Journal of Corporate Finance, 62, 101596.
- Brav, A., & Gompers, P. (2003). The role of lockups in initial public offerings. Review of Financial Studies, 16, 1–29.
- Carter, R., & Manaster, S. (1990). Initial public offerings and underwriter reputation. Journal of Finance, 45, 1045–1067.
- Chen, H., & Yang, S. (2020). Do investors exaggerate corporate ESG information? Evidence of the ESG momentum effect in the Taiwanese market. Pacific-Basin Finance Journal, 63, 101407.
- Chen, Y., Goyal, A., Veeraraghavan, M., & Zolotoy, L. (2020). Terrorist attacks, investor sentiment, and the pricing of initial public offerings. Journal of Corporate Finance, 65, 101780.
- Chourou, L., Saadi, S., & Zhu, H. (2018). How does national culture influence IPO underpricing? Pacific-Basin Finance Journal, 51, 318–341.
- Cornelli, F., Goldreich, D., & Ljungqvist, A. (2006). Investor sentiment and pre‐IPO markets. Journal of Finance, 61, 1187‒1216.
- Derrien, F. (2005). IPO pricing in “hot” market conditions: who leaves money on the table? Journal of Finance, 60, 487‒521.
- Dong, X., Meyer, J., Shmueli, E., Bozkaya, B., & Pentland, A. (2018). Methods for quantifying effects of social unrest using credit card transaction data. EPJ Data Science, 7, 1‒25.
- Duong, H., Goyal, A., Kallinterakis, V., & Veeraraghavan, M. (2022). Democracy and the pricing of initial public offerings around the world. Journal of Financial Economics, 145, 322‒341.
- Ellul, A., & Pagano, M. (2006). IPO underpricing and after-market liquidity. Review of Financial Studies, 19, 381–421.
- Engelen, P., & van Essen, M. (2010). Underpricing of IPOs: Firm-, issue- and country-specific characteristics. Journal of Banking and Finance, 34, 1958–1969.
- Espinosa-Méndez, C. (2022). Civil unrest and herding behavior: evidence in an emerging market. Economic Research-Ekonomska Istraživanja, 35, 1243–1261.
- Feng, Z., Chen, C., & Tseng, Y. (2018). Do capital markets value corporate social responsibility? Evidence from seasoned equity offerings. Journal of Banking and Finance, 94, 54–74.
- Hadzi-Vaskov, M., Pienknagura, S., & Ricci, L. (2023). The macroeconomic impact of social unrest. The B.E. Journal of Macroeconomics, Forthcoming.
- Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions, and organizations across nations. 2nd Edition. USA: Sage Publications.
- Kinnear, H., & Blanco, J. (2022). Cost of living crisis inflames civil unrest risks in emerging markets. https://www.maplecroft.com/insights/analysis/cost-of-living-crisis-inflames-civil-unrest-risks-in-emerging-markets/ (Accessed: May 31, 2023).
- Klomp, J. (2021). The Arab Spring and the international defense market. Business and Politics, 23, 202–220.
- Kollias, C., & Tzeremes, P. (2022). The nexus between social unrest and economic growth in Middle East and Central Asia countries. Review of Economics and Political Science, 7, 74–86.
- Leone, A., Rock, S., & Willenborg, M. (2007). Disclosure of intended use of proceeds and underpricing in initial public offerings. Journal of Accounting Research, 45, 111–153.
- Ljungqvist, A., Nanda, V., & Singh, R. (2006). Hot markets, investor sentiment, and IPO pricing. Journal of Business, 79, 1667‒1702.
- Ljungqvist, A., & Wilhelm, W. (2003). IPO pricing in the dot-com bubble. Journal of Finance, 58, 723–752.
- Loughran, T., & Ritter, J. (2004). Why has IPO underpricing changed over time? Financial Management, 33, 5–37.
- Marcato, G., Milcheva, S., & Zheng, C. (2018). Market integration, country institutions and IPO underpricing. Journal of Corporate Finance, 53, 87–105.
- Mazumder, S., & Saha, P. (2021). COVID-19: Fear of pandemic and short-term IPO performance. Finance Research Letters, 43, 101977.
- Redl C., & Hlatshwayo, S. (2021). Forecasting social unrest: A machine learning approach. IMF Working Paper.
- Ritter, J. (1984). The hot issue market of 1980. Journal of Business, 57, 215–240.
- Ritter, J. (1987). The costs of going public. Journal of Financial Economics, 19, 269–281.
- Rock, K. (1986). Why new issues are underpriced. Journal of Financial Economics, 15, 187–212.
- Schmeling, M. (2009). Investor sentiment and stock returns: Some international evidence. Journal of Empirical Finance, 16, 394–408.
- Sherman, A. (2000). IPOs and long-term relationships: An advantage of book building. Review of Financial Studies, 13, 697–714.
- Tanyeri, B., Savaser, T., & Usul, N. (2022). The stock and CDS market consequences of political uncertainty: The Arab Spring. Emerging Markets Finance and Trade, 58, 1821–1837.
- Welch, I. (1992). Sequential sales, learning and cascades. Journal of Finance, 47, 695–722.

### Appendix – Variable definitions (exact definitions as provided)
- Dependent variable
  - Underpricing: The first-day secondary market closing price divided by the IPO offer price, minus one. Sources: Datastream and SDC.
- Main independent variables
  - RSUI: The Reported Social Unrest Index, which is measured as the fraction of all articles considered that indicate social unrest in a particular market. We rebase the RSUI to an average of 100 for each market over our 1998-2018 sample period. Source: Barrett et al. (2022a).
- Primary control variables
  - Democracy: Index that measures the extent to which all social groups can participate in the political process. Source: V-Dem Institute.
  - Financial integration: Financial globalization index. Source: KOF Swiss Economic Institute.
  - Economic freedom: Economic freedom summary index. Source: Fraser Institute.
  - Market concentration: Measure of dispersion of trade value across an exporter’s partners. Source: World Bank.
  - Earnings management: The median ratio of total accruals over the lagged total assets for each market, measured over the five years preceding the IPO. Source: Bhattacharya et al. (2003).
  - IPO activity: Market-year ratio of the total number of IPOs divided by the number of publicly listed firms. Source: World Bank.
  - Market return: The return on the index for the listing market over the three months preceding the offering. Source: Datastream.
  - Liquidity: Market-level ratio of the total value of shares traded to aggregate market capitalization. Source: World Bank.
  - Offer size: The inflation-adjusted offer value in millions of USD. Source: SDC.
  - Top underwriter: Indicator variable set equal to one for IPOs underwritten by an investment bank in the top 25 of the league tables in the year of issuance. Source: SDC.
  - VC backed: Indicator variable set equal to one for IPOs that previously received VC funding. Source: SDC.
  - Lockup length: Days between the IPO issue date and the first lockup expiration date. Source: SDC.
  - Bookbuilt: Indicator variable set equal to one for bookbuilt IPOs. Source: SDC.
  - Firm commitment: Indicator variable set equal to one for firm commitment IPOs. Source: SDC.
  - Carve-out: Indicator variable set equal to one for equity carve-out IPOs. Source: SDC.
  - High tech: Indicator variable set equal to one for IPO firms in one of the high-tech SIC groupings identified by Ljungqvist and Wilhelm (2003). Source: SDC.
- Instruments
  - Neighbor unrest: Indicator variable set equal to one if a neighbor had an unrest event in the previous six months. Source: Barrett et al. (2022a).
  - Neighbor unrest (#): Number of neighbors that had an unrest event in the previous six months. Source: Barrett et al. (2022a).
  - Misery index: Sum of the unemployment rate and the inflation rate. Source: World Bank.
- Institutional quality measures (World Bank)
  - Control of corruption, Government effectiveness, Political stability, Regulatory quality, Rule of law, Voice and accountability.
- Limits to arbitrage measures (Boulton et al. (2020))
  - Short selling not allowed: Indicator variable set equal to one for IPOs issued where short selling is banned.
  - Security lending banned: Indicator variable set equal to one for IPOs issued where security lending is banned.
  - Short selling not practiced: Indicator variable set equal to one for IPOs issued where short selling is not routinely practiced.
- Additional return measures
  - Two-week return: The 10th-day secondary market closing price divided by the IPO offer price, minus one. Sources: Datastream and SDC.
  - One-month return: The 22nd-day secondary market closing price divided by the IPO offer price, minus one. Sources: Datastream and SDC.
  - Offer-to-open return: The return measured from the IPO offer price to the first-day opening price. Sources: Datastream and SDC.
  - Open-to-close return: The return measured from the first-day opening price to the first-day closing price. Sources: Datastream.

### Figures and Tables — key dataset and selected empirical estimates (exact reported numbers)
- Sample and descriptive scope
  - Sample of 15,145 IPOs issued from 1998-2018 (reported in Table 1 and Table notes).
  - Table 1 reports IPO market-level summary; example entries include:
    - US: IPOs 2,652; Average RSUI at IPO 87.89; Unrest Events 0; Underpricing 26.87%
    - Mainland China: IPOs 2,357; Average RSUI at IPO 115.28; Unrest Events 1; Underpricing 62.00%
    - UK: IPOs 1,349; Average RSUI at IPO 74.37; Unrest Events 5; Underpricing 16.01%
- Descriptive statistics (Table 2)
  - Underpricing: Unrest subsample N 184 Mean 0.233; No unrest subsample N 14,961 Mean 0.314; Difference -0.080 p-value 0.039.
  - RSUI: Unrest subsample Mean 637.012; No unrest subsample Mean 88.001; Difference 549.011 p-value 0.000.
  - Sample-level counts: Unrest observations 184; No unrest observations 14,961.
- Correlation matrix (Table 3)
  - Correlation of Underpricing with RSUI: -0.029 (reported).
  - Correlation of Underpricing with Market return: 0.150.
- IPO characteristics following unrest (Table 4, odds ratios)
  - Bookbuilt: Odds ratio 0.4483*** (IPO within three months of unrest event, Location FE).
  - Liquidity: Odds ratio 4.1936*** (IPO within three months of unrest event, Location FE).
  - Observations reported in columns: 8,313; 11,789; 15,145 (various specifications).
- Main regression results — RSUI and underpricing (Table 5)
  - RSUI coefficient: -0.0089** (Location FE), -0.0083** (Location FE alternative), -0.0052** (HLM), -0.0149*** (WLS).
  - Democracy coefficient examples: 0.3936***, 0.3900***, 0.2646**.
  - Offer size (log) coefficient: -0.0437*** (consistent across specifications).
  - Observations: 15,145; R-squared examples: 0.137, 0.144.
- Endogeneity / IV results (Table 6)
  - First-stage RSUI’ coefficients: -0.0867* and -0.1215*** (reported in First-stage / Second-stage columns).
  - Instruments: Neighbor unrest (0.2672***), Neighbor unrest (#) (0.2410***), Misery index (0.0069, 0.0089** in different columns).
  - Anderson χ2: 73.64 and 119.149 reported.
  - Cragg-Donald F-stat: 36.89 and 59.87; Stock-Yogo (10%) critical value 19.93.
  - Observations: 15,143; First-stage/Second-stage R-squared reported: 0.109 and 0.078.
- Institutional quality interactions (Table 7)
  - RSUI main effects across institutional measures (examples): Control of corruption RSUI -0.0345***; Government effectiveness RSUI -0.0556***; Political stability RSUI -0.0260***.
  - RSUI × Indicator interaction terms are positive and significant across indicators (e.g., 0.0193***, 0.0332***).
  - Observations: 15,145; Number of locations: 36.
- Limits to arbitrage (Table 8)
  - RSUI × Short selling measure coefficients: -0.0493*** (Short selling banned), -0.0485*** (Security lending banned), -0.0316*** (Short selling not practiced).
  - Short selling measure main effects: 0.2821***, 0.4444***, 0.0093.
  - Observations: 15,145; Number of locations: 36.
- Non-linearity (Table 9)
  - RSUI coefficient: 0.0156 (t-stat 1.54).
  - RSUI2 coefficient: -0.0040*** (t-stat -2.58).
  - Observations: 15,145; R-squared 0.137.
- Alternative return measures (Table 10)
  - RSUI coefficients for alternative dependent variables:
    - Two-week return: RSUI -0.0132** (t-stat -2.26).
    - One-month return: RSUI -0.0102* (t-stat -1.88).
    - Offer-to-open return: RSUI -0.0108*** (t-stat -3.40).
    - Open-to-close return: RSUI -0.0017 (t-stat -1.42).
  - Observations vary by measure: 15,143; 14,953; 14,642; 14,554.
- Robustness / influential locations (Table 11)
  - Exclusion regressions report RSUI coefficients across multiple exclusions; example coefficients:
    - Excluding Mainland China: RSUI -0.0060* (t-stat -1.73).
    - Excluding Hong Kong SAR: RSUI -0.0062* (t-stat -1.70).
    - Excluding U.S.: RSUI -0.0088** (t-stat -2.42).
    - Observations reported per exclusion: e.g., 12,788; 13,873; 13,796; 12,493; 15,121; 15,091; 14,967; 15,060; 8,621.

*Content extracted from the "References" and Appendix sections (including figures and tables summaries) of the PDF chapter "wpiea2023232-print-pdf - References".*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2023/english/wpiea2023232-print-pdf.pdf_