CrossRef citations to date
Original Articles

Misinformation, economic threat and public support for international trade

, & ORCID Icon
Pages 571-597
Published online: 09 Oct 2020


The recent surge in protectionist sentiment in countries around the world has rekindled the long-standing debate over the determinants of citizens’ trade policy preferences. We examine the influence of two understudied but increasingly relevant factors – misinformation and economic threat – on support for international trade in the United States. We first show that more than 6–in–10 Americans endorse a salient misperception about Chinese currency manipulation despite extensive evidence to the contrary. Then, based on a preregistered survey experiment, we show that misinformation can be corrected, regardless of whether the threatening frame is present or not. In contrast to these results on factual beliefs, however, we find that trade policy preferences are considerably stable: neither anti-trade misinformation nor an economically threatening frame significantly reduces support for international trade. These findings suggest that political elites’ strategy of ‘playing the China card’ by using misleading and threatening rhetoric is not so effective in mobilizing opposition to trade.


We thank Jamie Druckman, Jeffry Frieden, Alexandra Guisinger, Soo Yeon Kim, Katja Kleinberg, Philip Lipsy, Megumi Naoi, Katy Powers, and Megan Stewart for helpful comments. The data collection for this paper was partly funded by the Shorenstein Postdoctoral Fellowship at Stanford University.

Disclosure statement

No potential conflict of interest was reported by the author(s).


1 For example, see ‘Trump Wrong about WTO Record’, FactCheck.org, October 27, 2017, https://www.factcheck.org/2017/10/trump-wrong-wto-record/ (last accessed on August 12, 2020).

2 Although Guisinger (Citation2017, ch. 4, 7) reports several experiments that randomize exposure to corrective information about preexisting misperceptions (e.g. the widespread belief that China is the United States’ largest trading partner), her design does not randomize exposure to corrective information. Our design, which is the first to randomize exposure to misinformation and correction separately in the context of trade policy, allows us to isolate the individual and joint effects of both factors on trade beliefs and opinions.

3 Scholars have attempted to resolve these mixed results. For example, Jamal and Milner (Citation2019) show that even within import-competing sectors, individuals may have mixed trade preferences: those working in firms embedded in global value chains (GVCs) are less protectionist than those outside of GVCs.

4 In a related study, Hainmueller and Hiscox (Citation2006) argue that the observed relationship between educational attainment and pro-trade attitudes reflects the effect of exposure to economic ideas, such as the theory of comparative advantage, rather than the impact of skill levels.

5 A related, if not identical, concept is ‘priming effects’, which also refer to changes in opinions or evaluations resulting from increasing the salience of certain considerations. We follow previous research in using ‘framing effects’ to describe changes in policy preferences, which are the outcome of interest in our study (see, e.g. Busby et al., Citation2018).

6 The New York Times, February 4, 2010. ‘Currency Dispute Likely to Further Fray U.S.-China Ties’, https://nyti.ms/2mlZ1vx (last accessed on August 12, 2020).

7 The PBoC’s official website, June 19, 2010. ‘Further Reform the RMB Exchange Rate Regime and Enhance the RMB Exchange Rate Flexibility’, http://www.pbc.gov.cn/english/130721/2845862/index.html (last accessed on August 12, 2020).

8 Time.com, June 28, 2016. ‘Read Donald Trump’s Speech on Trade’, http://ti.me/292rlfV (last accessed on August 12, 2020).

9 CBS News, April 30, 2017. ‘President Donald Trump’s interview with ‘Face the Nation,’ https://www.cbsnews.com/news/trump-interview-full-transcript-face-the-nation/ (last accessed on August 12, 2020).

10 Reuters, August 22, 2018. ‘Exclusive: Trump demands Fed help on economy, complains about interest rate rises’, https://reut.rs/2MA6jL9 (last accessed on August 12, 2020).

11 For a comprehensive update on currency manipulation, see a report by the Peterson Institute for International Economics at https://piie.com/blogs/trade-investment-policy-watch/currency-manipulation-update-2015-17 (last accessed on August 12, 2020).

12 Brookings Institution, January 26, 2017. ‘China is Struggling to Keep Its Currency High, Not Low’, http://brook.gs/2kxl9QX (last accessed on August 12, 2020).

13 Treasury Department, August 5, 2019. ‘Treasury Designates China as a Currency Manipulator’, https://home.treasury.gov/news/press-releases/sm751(last accessed on August 12, 2020)

14 UBS, August 12, 2019. ‘What’s behind the ‘Currency Manipulator’ label?’ https://www.ubs.com/global/en/wealth-management/marketnews/home/article.1447198.html (last accessed on August 12, 2020).

15 The New York Times, August 5, 2019. ‘Trump’s China Shock’, https://nyti.ms/2T6zpCu (last accessed on August 12, 2020).

16 The targeted population is the adult population in the U.S. but participants are users on websites in the Google Surveys Publisher Network (see https://surveys.google.com/ for details). We use sampling weights Google Surveys estimated for our analysis. These weights are based on the distributions of sex (male, female), age groups (18–24, 25–34, 35–44, 45–54, 55–64, 65+), and regions (Midwest, Northeast, South, West) in the population. To follow IRB’s instructions, we added the following sentence at the end of this question: ‘(This is research. Your response is voluntary.)’

17 A unipolar scale is ideal for measuring factual beliefs since bipolar scales arguably fail to meet the mutually exclusive requirement for well constructed response scales (e.g. a claim that is somewhat true is by definition also somewhat untrue). There is an important methodological debate about the issues of measuring knowledge, misperceptions, or conspiracy beliefs (e.g. Clifford et al., Citation2019; Graham, Citation2020). Examining how the results could be different depending on alternative approaches is left for our future research.

18 ABC News, November 9, 2017. ‘10 times Trump Attacked China and Its Trade Relations with the U.S.’, http://abcn.ws/2nJMSQs (last accessed on August 12, 2020).

19 CNN News, May 2, 2016. ‘Trump: ‘We Can’t Continue to Allow China to Rape Our Country’’, http://cnn.it/1Z0VD4e (last accessed on August 12, 2020).

20 We preregistered the experiment with Experiments in Governance and Politics (https://egap.org/registration/2761).

21 Full survey instruments for both waves, together with all data and computer scripts, are included in a complete replication package (to be published at Dataverse).

22 The order of these outcomes measures was randomized.

23 Not all the effects we could estimate based on our 2 × 3 experimental design are reported, because we preregistered some of these effects given our theoretical expectations.

24 The baseline for the control group (i.e. no information) is 2.324. The coefficient estimate for the effect of misinformation vis-à-vis no information (without a threatening message) on beliefs is −0.333 and the standard error is 0.064. Therefore, the effect is highly significant. We did not pre-register this analysis because an important debate in the literature of misinformation is the effectiveness of correction (vis-à-vis misinformation), rather than the effect of misinformation per se compared to no information. Importantly, however, this significant effect, together with the significant effect of misinformation correction (Hypothesis 2), suggest that respondents were properly ‘manipulated’ in the directions we expect. These results suggest that the null effect on attitudes are likely to be neither the lack of statistical power nor the lack of respondents’ attention to our treatment materials.

25 Gallup, March 11, 2019. ‘Americans’ Favorable Views of China Take 12-Point Hit’, https://news.gallup.com/poll/247559/americans-favorable-views-china-point-hit.aspx (last accessed on August 12, 2020).

Log in via your institution

Log in to Taylor & Francis Online

PDF download + Online access

  • 48 hours access to article PDF & online version
  • Article PDF can be downloaded
  • Article PDF can be printed
USD 50.00 Add to cart

Issue Purchase

  • 30 days online access to complete issue
  • Article PDFs can be downloaded
  • Article PDFs can be printed
USD 308.00 Add to cart

* Local tax will be added as applicable

Related Research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.