Skip to main content

Advertisement

SpringerLink
Log in

The Far Reach of Hurricane Maria:

Spillover effects on U.S. Pharmaceutical Sectors and Other Exposed Industries

  • Original Paper
  • Published:
Economics of Disasters and Climate Change Aims and scope Submit manuscript

Abstract

Environmental degradation raises the frequency of natural disasters, and the growing reliance on global value chains exposes domestic labor markets to the ripple effects of these international calamities. To date, we know relatively little about such implications for U.S. labor markets. We leverage the significant disruption of Puerto Rican production and exports due to Hurricane Maria to study the spillover effects on employment in mainland U.S. labor markets. We find that the reduction in Puerto Rican import competition raises U.S. employment and the number of manufacturing establishments, particularly among pharmaceutical sectors with the highest level of industry exposure.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Code Availability

The code generated during the current study is available from the corresponding author on reasonable request.

Notes

  1. Exposure refers to the degree of import competition that U.S. industries and labor markets face from PR. The higher the degree of competition with Puerto Rican production prior to Hurricane Maria, the greater the level of industry exposure to the potential ripple effects post the disaster and disruption of Puerto Rican productions and exports. We provide more rigorous definitions of industry and labor market exposures in Section 2.

  2. A labor shed commonly refers to an area or region from which an employment center draws its commuting workers.

  3. We restrict the respective estimation samples to the top 25% of exposed industries and/or labor markets to strengthen our identification strategy and satisfy the parallel paths assumption underlying our DiD methodology.

  4. Over these first 8 months, electricity was restored gradually. Anecdotal evidence suggests that some factories were able to run at partial capacity shortly after Hurricane Maria using backup generators.

  5. Aside from PR, the U.S. imports pharmaceuticals from a number of European countries including Belgium, Denmark, France, Germany, Italy, the UK, the Netherlands, and Canada.

  6. While publicly available trade data published by the U.S. Census includes information on exports and imports of U.S. states and the territory of Puerto Rico to and from other foreign countries, the statistics do not include any information on U.S. mainland exports or imports to and from Puerto Rico directly.

  7. The replacement of U.S. pharmaceutical imports from PR with imports from other foreign countries potentially mitigates the effects of the disaster-induced reduction in Puerto Rican import competition on U.S. employment. In other words, without U.S. adjustments in import sourcing, one would expect the magnitude of our coefficient estimates to increase.

  8. In the same press release, the FDA announced collaboration with B. Braun and ICU Medical to address the amino acid shortage (FDA 2017d).

  9. Given the size of the Puerto Rican population relative to that of the U.S. mainland (≈ 1%), the aggregate changes in demand due to Hurricane Maria may be rather limited and are not the central focus of this study. However, to the extent that local demand changes mimic those at the aggregate level, our analysis of the overall local labor market effects of Hurricane Maria will encompass these changes.

  10. Hurricane Maria made landfall in September of 2017. Our data on employment, wages, and establishment counts are of quarterly frequency. As a result, Q3 of 2017 covers time before and after treatment. We opt to exclude Q3 of 2017 for our primary estimation and test the sensitivity of our results against this assumption. Reassuringly, our findings do not depend on this sample restriction (see Tables 6 and 7).

  11. We note that estimates of a linear treatment effect are robust to our primary approach (see Table 6, column (5)).

  12. In fact, preliminary analysis of the data demonstrates that alternative approaches comparing, for example, labor market outcomes at the bottom and top quantiles of industry and local exposures tend to violate the parallel paths assumption pre-treatment.

  13. We note that our results are robust to alternative thresholds as well as the interactions with the continuous exposure measures.

  14. As a robustness exercise to the inclusion of CZ fixed effects, we also explore alternative specifications including state-level fixed effects in combination with local labor market characteristics provided through the 2010 U.S. Census. Following the literature, we lag these local characteristics in an effort to avoid potential endogeneity of changing labor market conditions that contemporaneously adjust to the event and repercussions of Hurricane Maria. Our specific choice of labor market controls is guided by the previous literature and includes the percent of employment in manufacturing, the percent of the population that is college-educated, the percent of the population that is foreign-born, the percent of employment held by women (see ADH, 2013), and, particularly relevant to our study, the percent of the population of Puerto Rican descent. See our baseline estimates presented in Table 4.

  15. Controlling for industry- and commuting-zone-specific fixed effects precludes the inclusion of our exposure-based treatment indicators, which are time-invariant. Similarly time fixed effects control for common differences between pre- and post-treatment periods.

  16. Table 1 shows that CZ-industry employment and wages are widely dispersed. We test the sensitivity of our primary findings against the exclusion of employment, wage, and establishment count outliers. The restricted sample analysis produces quantitatively and qualitatively consistent results, which are available upon request.

  17. With respect to the location-specific control variables, we observe that labor markets with greater lagged shares of manufacturing employment, for example, are associated with smaller manufacturing employment today. Similarly, an increase in the percent of college-educated residents lowers CZ manufacturing employment, while a rise in the percent of foreign-born residents raises average commuting zone manufacturing employment. Similar to ADH (2013), we also find that the percentage of employment held among women is negatively correlated with average manufacturing-CZ employment.

  18. As is evidenced by the research of ADH (2013), Acemoglu et al. (2016), Hakobyan and McLaren (2016), and Dix-Carneiro and Kovak (2017), among others, changes in import competition tend to affect labor markets not just at the employment margin, but also in terms of wages, labor force participation, the number of establishments, and transfer payments.

  19. Of course, TCJA may also contribute to this employment growth in 2018.

  20. With respect to the location exposure effect, we also explore the issue of excluding non-manufacturing industries from the calculation of local exposure. Following the derivations by Kovak (2013), we test whether the inclusion of non-manufacturing industries in the determination of CLE biases our estimate of the local exposure effect as suggested by Kovak. We find some corroborating evidence that the inclusion of non-manufacturing employment – for which the RCA is zero – in the calculation of cumulative local exposure understates the changes in local labor demand and consequently overstates the importance of the Puerto Rican trade shock to U.S. mainland labor markets. The biased coefficient estimate of the direct local exposure effect on employment increases slightly and switches signs with respect to wages and establishment counts. Nonetheless, each of these coefficient estimates is rather imprecisely estimated and remains statistically insignificant (see column (3) of Table 7 in the Appendix).

  21. The sample plotted in Figs. 5a through d includes all industries in the top quantile of industry exposure, i.e. RCAj > 75th percentile

  22. The sample plotted in Figs. 6a through d includes all commuting zones in the top quantile of cumulative location exposure, i.e. CLEc > 75th percentile

References

  • Acemoglu D, Autor D, Dorn D, Hanson GH, Price B (2016) Import competition and the great us employment sag of the 2000s. J Labor Econ 34(S1):S141–S198

    Article  Google Scholar 

  • Alexander M, Zagheni E, Polimis K (2019) The impact of hurricane maria on out-migration from puerto rico: Evidence from facebook data

  • Bernard AB, Jensen JB, Redding SJ, Schott PK (2018) Global firms. J Econ Lit

  • Autor DH, Dorn D, Hanson GH (2013) The China syndrome: Local labor market effects of import competition in the united states. Am Econ Rev 103(6):2121–68

    Article  Google Scholar 

  • Autor DH, Dorn D, Hanson GH, Song J (2014) Trade adjustment: Worker-level evidence. Q J Econ 129(4):1799–1860

    Article  Google Scholar 

  • Autor DH, Dorn D, Hanson GH (2016) The China shock: Learning from labor-market adjustment to large changes in trade. Ann Rev Econ 8:205–240

    Article  Google Scholar 

  • Belasen AR, Polachek SW (2008) How hurricanes affect wages and employment in local labor markets. Am Econ Rev 98(2):49–53

    Article  Google Scholar 

  • Blanchard OJ, Katz LF, Hall RE, Eichengreen B (1992) Regional evolutions. Brookings Pap Econ Act 1992(1):1–75

    Article  Google Scholar 

  • Boehm CE, Flaaen A, Pandalai-Nayar N (2019) Input linkages and the transmission of shocks: Firm-level evidence from the 2011 tōhoku earthquake. Rev Econ Stat 101(1):60–75

    Article  Google Scholar 

  • Bown C (2020) Us-china trade war tariffs: An up-to-date chart. Peterson Insitute for International Economics

  • Browning TN, Sawyer DE, Brooks GR, Larson RA, Ramos-Scharrón C. E, Canals-Silander M (2019) Widespread deposition in a coastal bay following three major 2017 hurricanes (irma, jose, and maria). Sci Rep 9(1):7101

    Article  Google Scholar 

  • Cabán P (2019) Hurricane maria’s aftermath: redefining Puerto Rico’s colonial status

  • Cavallo E, Galiani S, Noy I, Pantano J (2013) Catastrophic natural disasters and economic growth. Rev Econ Stat 95(5):1549–1561

    Article  Google Scholar 

  • Dix-Carneiro R, Kovak BK (2015) Trade liberalization and the skill premium: A local labor markets approach. Am Econ Rev 105(5):551–57

    Article  Google Scholar 

  • Dix-Carneiro R, Kovak BK (2017) Trade liberalization and regional dynamics. Am Econ Rev 107(10):2908–46

    Article  Google Scholar 

  • Dix-Carneiro R, Kovak BK (2019) Margins of labor market adjustment to trade. J Int Econ 117:125–142

    Article  Google Scholar 

  • Emanuel K (2005) Increasing destructiveness of tropical cyclones over the past 30 years. Nature :436

  • FDA (2017a) Statement from fda commissioner scott gottlieb, m.d., on fda’s continued assistance following the natural disaster in puerto rico. https://www.fda.gov/news-events/press-announcements/statement-fda-commissioner-scott-gottlieb-md-fdas-continued-assistance-following-natural-disaster

  • FDA (2017b) Securing the future for puerto rico: Restoring the island’s robust medical product manufacturing sector. https://www.fda.gov/media/108975

  • FDA (2017c) Testimony of scott gottlieb, m.d. commissioner of food and drugs; examining hhs’s public health preparedness for and response to the 2017 hurricane season. https://www.fda.gov/news-events/congressional-testimony/examining-hhs-public-health-preparedness-and-response-2017-hurricane-season-10242017

  • FDA (2017d) Statement by fda commissioner scott gottlieb, m.d., updating on puerto rico related medical product shortages. https://www.fda.gov/news-events/press-announcements/statement-fda-commissioner-scott-gottlieb-md-updating-puerto-rico-related-medical-product-shortages

  • FDA (2018) Statement by fda commissioner scott gottlieb, m.d., update on recover efforts in puerto rico, and continued efforts to mitigate iv saline and amino acid drug shortages. https://www.fda.gov/news-events/press-announcements/statement-fda-commissioner-scott-gottlieb-md-update-recovery-efforts-puerto-rico-and-continued

  • Felbermayr G, Gröschl J, Heid B (2018) The short-run trade effects of natural disasters

  • Feliciano ZM (2018) Irs section 936 and the decline of puerto rico’s manufacturing. Cent J 30(3)

  • Fowler CS, Rhubart DC, Jensen L (2016) Reassessing and revising commuting zones for 2010: History, assessment, and updates for us ‘labor-sheds’ 1990–2010. Popul Res Policy Rev 35(2):263–286

    Article  Google Scholar 

  • Friedt F (2018) Natural disasters, trade and local factor prices: Labor market externalities arising from the disaster-induced diversion of trade

  • Friedt F (forthcoming) Natural disasters, aggregate trade resilience and persistent local disruptions: Evidence from hurricane katrina. Rev Int Econ

  • Fujita M, Mori T (1996) The role of ports in the making of major cities: self-agglomeration and hub-effect. J Dev Econo 49(1):93–120

    Article  Google Scholar 

  • Gassebner M, Keck A, Teh R (2010) Shaken, not stirred: the impact of disasters on international trade. Rev Int Econ 18(2):351–368

    Article  Google Scholar 

  • Glaeser EL, Gyourko J (2005) Urban decline and durable housing. J Polit Econ 113(2):345–375

    Article  Google Scholar 

  • Hakobyan S, McLaren J (2016) Looking for local labor market effects of nafta. Rev Econ Stat 98(4):728–741

    Article  Google Scholar 

  • Hinojosa J, Meléndez E (2018) Puerto rican exodus: One year since hurricane maria. Center for Puerto Rican Studies

  • Keellings D, Ayala JJH (2019) Extreme rainfall associated with hurricane maria over Puerto Rico and its connections to climate variability and change. Geophys Res Lett 46(5):2964–2973

    Article  Google Scholar 

  • Kirchberger M (2017) Natural disasters and labor markets. J Dev Econ 125:40–58

    Article  Google Scholar 

  • Kishore N, Marqués D., Mahmud A, Kiang MV, Rodriguez I, Fuller A, Ebner P, Sorensen C, Racy F, Lemery J, Mass L, Leaning J, Irizarry RA, Balsari S, Buckee CO (2018) Mortality in Puerto Rico after hurricane maria. New Engl J Med 379(2):162–170

    Article  Google Scholar 

  • Kovak BK (2013) Regional effects of trade reform: What is the correct measure of liberalization?. Am Econ Rev 103(5):1960–76

    Article  Google Scholar 

  • Lim D (2019) B. braun invests $1b in iv fluid manufacturing to alleviate shortages. MedTechDive. https://www.medtechdive.com/news/b-braun-invests-1b-in-iv-fluid-manufacturing-to-alleviate-shortages/554855/

  • Martincus CV, Blyde J (2013) Shaky roads and trembling exports: Assessing the trade effects of domestic infrastructure using a natural experiment. J Int Econ 90(1):148–161

    Article  Google Scholar 

  • McIntosh MF (2008) Measuring the labor market impacts of hurricane katrina migration: Evidence from Houston, Texas. Am Econ Rev 98(2):54–57

    Article  Google Scholar 

  • Meléndez-Vazquez F, Olmeda-Saldaña M, Cruz J, Arcila D, Betancur R (2019) Effects of hurricane maria in hamlet communities (serranidae Hypoplectrus spp.) in Puerto Rico. Ecol Indic 105591:107

    Google Scholar 

  • Miller PW, Kumar A, Mote TL, Moraes FDS, Mishra DR (2019) Persistent hydrological consequences of hurricane maria in puerto rico. Geophys Res Lett 46(3):1413–1422

    Article  Google Scholar 

  • Mora MT, Dávila A, Rodríguez H (2018) Migration, geographic destinations, and socioeconomic outcomes of puerto ricans during la crisis boricua: Implications for island and stateside communities post-maria. Cent J 30(3):208–229

    Google Scholar 

  • Moutray C (2019) A hiring engine: a breakdown of the job openings in manufacturing. National Association of Manufacturers

  • Noy I (2009) The macroeconomic consequences of disasters. J Dev Econ 88(2):221–231

    Article  Google Scholar 

  • Chang Hoon O h, Reuveny R (2010) Climatic natural disasters, political risk, and international trade. Glob Environ Chang 20(2):243–254

    Article  Google Scholar 

  • Perfecto I, Hajian-Forooshani Z, Iverson A, Irizarry AD, Lugo-Perez J, Medina N, Vaidya C, White A, Vandermeer J (2019) Response of coffee farms to hurricane maria Resistance and resilience from an extreme climatic event. Sci Rep 9(1):1–11

    Article  Google Scholar 

  • Ramos-Scharrón CE, Arima E (2019) Hurricane maría’s precipitation signature in Puerto Rico: A conceivable presage of rains to come. Sci Rep 9(1):1–7

    Article  Google Scholar 

  • Román MO, Stokes EC, Shrestha R, Wang Z, Schultz L, Carlo EAS, Sun Q, Bell J, Molthan A, Kalb V et al (2019) Satellite-based assessment of electricity restoration efforts in puerto rico after hurricane maria. PloS one 14(6)

  • Skidmore M, Toya H (2002) Do natural disasters promote long-run growth? Econ Inq 40(4):664–687

    Article  Google Scholar 

  • Strobl E (2012) The economic growth impact of natural disasters in developing countries: Evidence from hurricane strikes in the central american and caribbean regions. J Dev Econ 97(1):130–141

    Article  Google Scholar 

  • Sytsma T (2018) Do domestic trade frictions have global consequences? the welfare impact of hurricanes activity around us ports. The Welfare Impact of Hurricanes Activity Around US Ports (November 27, 2018)

Download references

Funding

Not applicable

Author information

Authors and Affiliations

  1. Department of Economics, Macalester College, 1600 Grand Avenue, Saint Paul, MN, 55105, USA

    Felix L. Friedt

  2. M.A. Clinical Mental Health Counseling, Regis University, 3333 Regis Blvd, Denver, CO, 80221, USA

    Abigail Crispin

Authors
  1. Felix L. Friedt
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Abigail Crispin
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Felix L. Friedt.

Ethics declarations

Conflict of Interest/Competing Interests

Not applicable

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Earlier versions of this paper were presented at the Macalester Economics Research Seminar and the Multinational Corporations Capstone Seminar. We thank the seminar participants, and in particular Sarah West, Amy Damon, Gabriel Lade, Lisa Mueller, Mario Solis-Garcia, Nahla Almbaid, Muxue Kou, Yannick, Laurent, Yuanchen Li, Selma Osman, Karl Schuetzle, Venkat Somala, Jennifer Tan, and Daijiro Yokota for their valuable comments. Any remaining errors are the authors’ alone.

Appendix:

Appendix:

Following the theoretical derivations by ADH (2013), which go beyond the scope of this paper, one can approximate the cumulative local labor market exposure as a function of the weighted sum of industry-specific Puerto Rican exports to the U.S. (Mpujt) relative to the size of the local labor market in manufacturing industries (Lit) and weighted by the share of local industry-specific employment (Lijt) relative to total U.S. employment in that particular industry (Lujt):

$$ IPW_{uit}= \underset{j=1}{\sum} \frac{L_{ijt}}{L_{ujt}}\frac{M_{pujt}}{L_{it}} $$
(5)

As the authors discuss, this measure of local exposure may be subject to several concerns regarding the anticipatory adjustments in labor market composition and potential endogeneity between foreign exports and U.S. industry employment due to common demand shocks. In response, the authors propose an instrumental variable approach, whereby labor market composition is fixed pre-treatment (\(\bar {t}\)) and foreign-country exports to the U.S. are instrumented for via foreign-country exports to other developed countries (Mpojt):

$$ IPW_{oit}=\underset{j=1}{\sum} \frac{L_{ij\bar{t}}}{L_{uj\bar{t}}}\frac{M_{pojt}}{L_{i\bar{t}}} $$
(6)

Applying this local exposure measure in the context of our study raises a few issues that require modifications from the original ADH (2013) specification. As previously discussed, our analysis focuses on the effects of the unexpected event of Hurricane Maria, rather than the anticipated joining of China in the WTO. Consequently, we fix the labor market composition one year prior to the event, rather than a full decade. Moreover, in the absence of data for Puerto Rican exports to the United States, we simply approximate U.S. imports from Puerto Rico with Puerto Rican exports to the rest of the world, rather than employing the 2-stage IV estimator.

Footnote 21

Fig. 5
figure 5

Labor market and Trade variations with Industry Exposure (RCAj > 75th perc)21

Full size image
Fig. 6
figure 6

Labor market and import competition variations with Location Exposure (CLEc > 75th perc)22

Full size image

Footnote 22

Table 7 Robustness - location exposure effects with alternative measures of exposure & sample restrictions
Full size table
Table 8 Robustness - change in employment (2016 Q3 & Q4 to 2017 Q3 & Q4)
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Friedt, F.L., Crispin, A. The Far Reach of Hurricane Maria:. EconDisCliCha 6, 29–71 (2022). https://doi.org/10.1007/s41885-021-00097-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s41885-021-00097-0

Keywords

Search

Navigation