Nikola: and Hindenburg Report Trust and Investor Perception

Nikola: and Hindenburg Report Trust and Investor Perception

• 1. © The Author(s) 2024 193 M. Born, Building Trust in Startup Communication, Springer Business Cases, https://doi.org/10.1007/978-3-031-63284-6_7 Chapter 7 The Hindenburg Report as a Turning Point 7.1 The Popularity of Search Queries Using Google Trends Prior to the publication of the Hindenburg report Nikola: How to Parlay An Ocean of Lies Into a Partnership With the Largest Auto OEM in America, Nikola was a highly successful new venture (Fig. 4.1). The Nikola share price was undoubtedly very volatile in the 3  months after its NASDAQ debut, but Trevor Milton was remarkably successful in convincing partner companies and investors of his company’s prospects. Nikola succeeded in integrating prominent partner companies into what the author considers a conceptionally well-designed vertical value chain (cf. Sect. 5.2.2). The Arizona-based startup was also exceptionally successful in the capital markets. Many retail investors had bought into a company that had virtually no revenue, causing Nikola’s market capitalization to briefy surpass even that of Ford shortly after going public (cf. Sect. 4.3). However, Nikola’s success was by no means limited to individual, non-professional investors; in fact, Nikola also managed to gain the trust of several reputable institutional investors, funds, and fnancial analysts. This chapter aims to investigate whether any change in public perception can be detected in the period after the short seller attack compared to the period before. First, it should be noted that, even before the publication of the said report on September 10, 2020, Nikola received public attention far beyond the competitive environment. Search queries for the term “Nikola Corporation” in Google Search across the United States peaked during Nikola’s remarkable NASDAQ debut in early June 2020, according to Google Trends (Fig. 7.1). In the wake of the short seller attack, a second peak occurred in mid-September 2020—one that was slightly less pronounced than the one in early June but longer lasting. Searches were particularly frequent in the state of Arizona, which is not surprising since Nikola is based there. The picture is quite different when one looks at the search queries for the term “Trevor Milton.” These remained at a relatively low level throughout the year 2020,

• 2. 194 Fig. 7.1 Interest over time and by subregion for the search query “Nikola Corporation” according to Google Trends except for September, when they suddenly rose sharply (Fig.  7.2). The increase coincided with the publication of the Hindenburg report, and searches peaked during the week of Trevor Milton’s resignation, when the frst allegations of sexual misconduct were made public. The search queries were particularly pronounced in the state of Utah, where Trevor Milton was born, attended college, and founded Nikola. In the context of Nikola’s NASDAQ debut in June 2020, public attention to the startup surged as expected, with public interest seeming to focus primarily on the company rather than the entrepreneur behind it. This changed dramatically in September. The analysis with Google Trends can be interpreted to mean that the short seller’s attack led to increased awareness of Trevor Milton in particular. As explained in Sect. 6.1.1, p. 163 and Sect. 6.1.6, Hindenburg heavily targeted Nikola founder Trevor Milton in its attack, and it can be concluded that this personal attack and the events that immediately followed were indeed noticed by a broader public and attributed not only to the startup Nikola but above all to the entrepreneur Trevor Milton. The following section will further examine whether and how reporting on Nikola and Trevor Milton changed after the publication of the short seller report. 7 The Hindenburg Report as a Turning Point

• 3. 195 Fig. 7.2 Interest over time and by subregion for the search query “Trevor Milton” according to Google Trends 7.2 Coverage of the Nikola Case in Selected Media The infuence of the media on the fnancial industry and in particular on investment decisions is multifaceted, and different research streams have reached different conclusions depending on their point of view. Carlo Raimondo (2019) has mapped the various research streams on the role of the media in fnance by dividing this universe of knowledge into two substreams that refect the two main areas of fnance: the direct impact of the media on fnancial markets in terms of asset prices, on the one hand, and the broader impact of the media on corporate fnance and governance, on the other hand (Raimondo, 2019, pp. 156–157). As far as the infuence of the media on stock price formation is concerned, there are streams that emphasize the positive contribution of the media in situations of asymmetric information, while others conclude that the media tends to increase investor irrationality, be it through unfounded enthusiasm or through sensationalism (Raimondo, 2019, pp. 158–160). Among others, business journalists refer to fnancial analysts, whose recommendations undoubtedly infuence stock markets (Whitehouse, 2022, p. 46; Whitehouse et al., 7.2 Coverage of the Nikola Case in Selected Media

• 4. 196 2018). In both the Nikola and Tesla cases, the performance of fnancial analysts has been largely poor during the period covered by this book, albeit in opposite directions in each case. While the vast majority of fnancial analysts have underestimated the performance of the Tesla share over a long period of time, they have massively and continuously overestimated the value of the Nikola share. Although fnancial analysts lowered the average price target of the Nikola share by around 50% between the beginning of August and the end of December 2020, the average price target was still signifcantly higher than the Nikola share price of the time. As of August 5, 2020, the average analyst target price for NKLA was $55 USD, according to Dow Jones/Barron’s, representing an upside of 62% (Root, 2020a). As of September 21, the average price target was lower, at $47.5 USD, but it promised a massive upside potential of over 75% (Assis, 2020). By December 23, 2020, the average analyst price target had dropped to $29 USD per share, with 3 of 8 analysts still rating the Nikola stock a buy (Root, 2020b). This implied an upside of nearly 100% (Nikola’s closing share price was $15.03 USD on December 23, 2020). For comparison, the closing price was $9.87 USD 1 year later, on December 31, 2021. Thus, while fnancial analysts were projecting a potential doubling of Nikola’s stock price at the end of 2020, the price fell 50% in the following 12 months against a growing NASDAQ overall market, raising questions about incentive structures within the expert group of fnancial analysts that are beyond the scope of this study. This analysis does not address the question of whether media coverage of the Nikola case may have infuenced corporate governance decisions in a particular direction. Nor is it intended to assess whether the media coverage of Nikola infuenced the share price in favor of market effciency or, conversely, whether it might have helped to promote irrational decision-making. This study merely assumes that media coverage affects stock market performance and that the stock price is an indicator of investor confdence in a company, although exogenous factors must be considered (cf. Sect. 2.1.3). In our case, the question of whether the publication of the Hindenburg report changed the topics, focus, and thus the general tone of media coverage of Nikola in any way is of particular interest. To answer this more precisely, the analysis used corpus linguistic methods and an open-source software, AntConc. Actual sentiment analysis was not conducted, despite the availability of specifc word lists for the realm of fnance—in such an analysis, a target corpus is matched with a word list as a reference corpus (cf. Loughran & McDonald, 2011). The reason is that, in a relatively narrow corpus, negations, or modifcations, such as amplifcations and attenuations, might have distorted the results in a considerable way. Nevertheless, a comparison of the collections of articles across two time periods does reveal insights. For this analysis, I collected all the English language articles in the Factiva database that had Nikola or Trevor Milton as their main topic and appeared in Reuters, Financial Times, The Wall Street Journal, Barron’s, and Forbes publications in the 4 months before and after the release of the Hindenburg Research report. I chose the 4 months before and after the event as the observation period to study a suffciently long period during which Nikola, as a publicly traded company, was obligated to provide ongoing public communications. As for the source selection, it was important to represent Reuters, one of the world’s largest news agencies, and to include 7 The Hindenburg Report as a Turning Point

• 5. 197 four specialized media products from the feld of business and fnancial reporting. To qualify for inclusion in the corpus, each article had to contain “Nikola” or “Trevor Milton” in the title.1 The resulting corpus included 279 articles—97 of these articles (128 articles minus 31 duplicates) were published before the Hindenburg report, and 182 of the articles (281 articles minus 99 duplicates) were published on and in the 4 months following the release date, September 10, 2020. The frst thing to notice is that the number of articles published in the 4 months following the short seller attack was almost double the number published in the previous period. This is remarkable in that Nikola’s NASDAQ debut, a landmark event for the startup, occurred early in the frst 4-month period. Also, the word clouds of keywords automatically generated in Factiva along with the search queries seem to indicate that the main topics of media coverage shifted signifcantly. However, these results should be interpreted with caution, as detailed information on the creation of these word clouds could not be found on the Dow Jones Factiva website. In its Factiva user manual, Dow Jones, the producer of the news database, writes: “Keywords display the terms and phrases with signifcant volume,” and “the keyword font size indicates the frequency of the term in the news for the time period selected” (Dow Jones and Company, 2012, p. 65). According to the University at Buffalo Libraries, the visualization generated by Factiva shows the top keywords examining the top 100 articles of the search (Klotzbach-Russell, 2022). Nevertheless, the keywords generated in this way give an indication of the most important topics covered in the respective periods. For the 4 months preceding the Hindenburg report, the main keywords related to Nikola’s entrepreneurial orientation and the industry in general (electric truck startup, battery-powered semi, electric vehicle, fuel cell, electric pickup). In addition, keywords that refer directly to the reverse merger in the context of Nikola’s public offering were listed (equity stake, special-purpose acquisition company) (Fig. 7.3, left). 1 In the Companies section of Factiva “Nikola Corp.” was selected, other search settings: Duplicates = Identical. Exclude = Republished news, recurring pricing, and market data, Web News. Fig. 7.3 Word clouds by the Factiva database for the Nikola target corpus from the 4 months before [left] and after the short seller report was published 7.2 Coverage of the Nikola Case in Selected Media

• 6. 198 The keywords generated in the 4 months following the Hindenburg report present a much different picture. The most prominent term in this later period—“short seller”—did not appear in the previous period. In addition, terms with a legal meaning appeared (intricate fraud, legal action). The fact that “marketing video” also appeared as a keyword underscores the importance that this element of the short seller attack assumed in media coverage. To generate a comparison, I applied the same approach to articles with the terms “Tesla” and/or “Elon Musk” in the title. The search query yielded a total of 759 articles that appeared in the selected media outlets during the 8 months relevant to this analysis. Of these, 305 (519 minus 214 duplicates) were published in the 4 months before September 10, 2020, and 454 articles (690 minus 236 duplicates) were published in the 4 months after. The automatically generated keyword display in Factiva provides very limited insights into the tone of media coverage. Nevertheless, it was noticeable that, unlike in the case of Nikola, in the case of Tesla, Inc. the word clouds were not remarkably different. The most important keywords described Tesla’s industry. Among them are terms like “electric vehicle maker” or “electric car maker.” Metals, which are important for the production and improvement of lithium-ion batteries, were also an important topic. While there were differences between the two-word clouds, there was no indication of a fundamental shift in themes (Fig. 7.4). One noteworthy and explainable difference concerns the reopening of a Tesla plant in Fremont, California in the frst time period, which directly followed the pandemic-induced dip in share prices in March 2020— this seems to have been an aspect of the media coverage in this frst period. The considerable rise in the Tesla share price also attracted media attention, albeit less than might have been expected (cf. Sect. 4.3). For further analysis with the corpus tool AntConc, I manually cleaned the PDF fles containing the Nikola article collections2 and converted them to .txt fles. The 2Factiva-generated source names, search summaries, and article classifcations that could distort the results have been manually removed. Fig. 7.4 Word clouds by the Factiva database for the Tesla reference corpus from the 4 months before [left] and after the short seller report was published 7 The Hindenburg Report as a Turning Point

• 7. 199 goal was to identify patterns in reporting from a distance, so to speak, that are diffcult to discern from a mere reading of various articles. To fnd out which words can be considered characteristic of the coverage in the 4 months after the publication of the Hindenburg report (target corpus) compared to the 4  months before (reference corpus), I used the keyword tool. As AntConc’s developer, Laurence Anthony (2022), writes, “this tool shows words that appear unusually frequently in the target corpus in comparison with the words in the reference corpus based on a statistical measure (i.e., ‘keywords’).”3 Figure 7.5 shows the 3 Indexer = simple_word_indexer; sort by likelihood. When using log likelihood as the statistical measure, the following signifcance values apply (Anthony, 2012): 95th percentile; 5% level; p < 0.05; critical value = 3.84 99th percentile; 1% level; p < 0.01; critical value = 6.63 99.9th percentile; 0.1% level; p < 0.001; critical value = 10.83 99.99th percentile; 0.01% level; p < 0.0001; critical value = 15.13. Fig. 7.5 Disproportionately frequent words in the 4-month period after the short seller report was published 7.2 Coverage of the Nikola Case in Selected Media

• 8. 200 Fig. 7.6 Words that appeared frequently within a distance of fve words to the left and right of the term “Hindenburg” 25 characteristic words for the period from September 10, 2020, to January 10, 2021. This word list provides an indication of the extent to which the Hindenburg report dominated coverage in the media outlets studied in the 4 months following its publication. It is a cluster of interconnected topics that have made it through the media flter. “Hindenburg” was the term that occurred most disproportionately (unusually) frequently in the period studied. It appeared 345 times, while it did not appear at all in the previous period. Figure 7.6 shows which words co-occurred with “Hindenburg” in the target corpus. Of these fve words, three were among the top six most disproportionately frequent words: “Short,” “seller,” and “report.” It is also not surprising that “research” and “released” were often mentioned in connection with “Hindenburg.” The full company name is Hindenburg Research, and as is well known, it was this short seller that released the critical report on Nikola. Applying the collocation tool to the term “mr,” which ranked fourth in the list of disproportionately frequent words in the keyword analysis, I found that “Milton” came frst by a wide margin, followed by Mark Russell, the then CEO of Nikola, and Nathan Anderson, the founder of Hindenburg Research (Fig. 7.7). Steve Girsky, Trevor Milton’s successor as Nikola chairman only came in fourth place. In ffth place was Jeff Ubben, founder of the investment company ValueAct, which made an early investment in Nikola (Sect. 5.3.3, p. 145). The Financial Times wrote about the fourth and ffth named executives in an article titled “Nikola: the clues in Trevor Milton’s past that investors missed or ignored”: Some investors in Nikola say they overlooked Mr Milton’s brashness because of the other executives backing the company—including Jeff Ubben, the founder of activist hedge fund ValueAct and Mr Girsky, a respected automotive executive. (P. Campbell et al., 2020) 7 The Hindenburg Report as a Turning Point

• 9. 201 Fig. 7.7 Words that appear frequently within three words to the right of the term “mr” This result confrms the important role that arguments from authority played in Nikola’s trust building described in Sect. 5.3.3. The patterns on the language surface of media coverage are consistent with the results that emerged from examining search queries in Google Trends: The short seller attack not only dominated the news coverage but was also directly associated with founder Trevor Milton. What is also noticeable is the accumulation of terms that can be associated with legal disputes in the broadest sense. These include terms such as “fraud,” “allegations,” “false,” “justice,” “misleading,” “misled,” but also “diligence,” “lawsuit,” or “legal.” The especially high number of mentions of the term “video” (91 mentions compared to zero mentions in the previous period) illustrates the role that this aspect of Hindenburg’s argumentatively relevant components played in media perception. The more substantial aspect of the business model from the point of view of this book was Trevor Milton’s false claim, conceded by Nikola itself, that the startup had managed to drastically reduce the cost of hydrogen production (cf. Sect. 5.2.2, p.  132). In contrast, the video argument was apparently more engaging and illustrative on the one hand while simultaneously undermining multiple strands of Trevor Milton’s and Nikola’s trust argument on the other (cf. Sect. 6.1.4). Numerous terms in the keyword analysis indicate that the period following the publication of the Hindenburg report was dominated by argumentation, as described by the qualitative case reconstruction. Keywords like “allegations,” “claims,” but also “accusations” and “accused” were strongly represented. For example, the word “rebuttal” never occurred in the previous period but occurred 27 times after the short seller attack (rank 61 of disproportionately frequent words). “Rebuttal” was frequently mentioned in connection with “refuting,” and “denied” also appeared in the list of the 74 disproportionately frequent words. All of these terms indicate the importance of aspects of counter argumentation (Sect. 2.3.3). Applying AntConc’s Key-Word-In-Context tool to the word “allegations” (rank 5 of the keyword list) reveals a strong tie with the term “fraud,” which in turn is closely linked to the terms “short” and “seller” (i.e., the source of the allegations). This confrms the impression that there is an interconnected cluster of topics that is dominant and directly related to the debate unleashed by the short seller attack (Fig. 7.8). 7.2 Coverage of the Nikola Case in Selected Media

• 10. 202 Fig. 7.8 How the word “allegations” was commonly used in the target corpus (KWIC Tool) Swapping the selected target and reference corpora reveals words that occur disproportionately less frequently in the 4 months after September 10, 2020, than in the period before (Fig. 7.9). The list shows that, in the previous period, Nikola was much more often associated with Tesla (analogy or contrast). The same applies, albeit somewhat less emphatically, to Elon Musk (44 references in the preceding period vs. 23 thereafter). Unsurprisingly, terms directly related to Nikola’s NASDAQ debut in early June 2020 appeared (“IPO” and “SPAC”), as well as words related to fnancial reporting (“stocks” and “warrants” and “market”) or the industry and products in general (“electric” and “cybertruck”). The word “traditional” (rank 16) occurs particularly frequently in conjunction with “auto” and “maker,” suggesting that Trevor Milton’s battle against the old guard of OEM’s did make its way into the media in the preperiod, but that there was no longer much use for such components of Trevor Milton’s storytelling repertoire in the post short seller attack period (cf. Sects. 5.1.1 and 5.1.2). Much the same can be said about the Tesla comparison. Although Trevor Milton increasingly distanced himself from Tesla and Elon Musk the more Nikola gained prominence in the public perception, Tesla has long remained an important narrative containing an argumentative core in the context of Nikola’s rise (Sect. 4.2, p. 93, Sect. 5.2.2, p. 133, Sect. 5.2.3, p. 138, Sect. 5.3.3, p. 149). 7 The Hindenburg Report as a Turning Point

• 11. 203 Fig. 7.9 Words that appeared disproportionately less frequently in the 4-month period after the short seller report was published References Anthony, L. (2012). AntConc (Windows, Macintosh OS X, and Linux). Build 3.3.5. Retrieved from http://www.laurenceanthony.net/software/antconc/releases/AntConc335/help.pdf Anthony, L. (2022). AntConc (Windows, MacOS, Linux). Build 4.0.4. Retrieved from https://www. laurenceanthony.net/software/antconc/releases/AntConc404/help.pdf Assis, C. (2020, September 21). Here’s what Wall Street is saying about Nikola founder’s ‘shocking’ departure; Founder’s exit could ‘minimize the drama,’ analyst says. MarketWatch. References

• 12. 204 Retrieved from http://global.factiva.com/redir/default.aspx?P=sa&an=MRKWC00020200921 eg9l005pl&cat=a&ep=ASE Campbell, P., Bushey, C., & Ortenca, A. (2020, December 7). The clues investors missed or ignored; FT BIG READ. NIKOLA allegations about Trevor Milton’s business practices and personal life continue to raise due diligence concerns in his truck start-up Nikola, whose share price has fallen over 70 per cent since it listed in June. Financial Times. Retrieved from https://advance. lexis.com/api/document?collection=news&id=urn:contentItem:61FY-C501-DYTY-C05M-00 000-00&context=1516831 Dow Jones and Company. (2012). Factiva-User-Guide_EN.pdf (p. 111). Retrieved from https:// ethz.ch/content/dam/ethz/associates/ethlibrary-dam/documents/Standorteundmedien/ Medientypen/Datenbanken/Factiva/Factiva-User-Guide_EN.pdf Klotzbach-Russell, C. (2022, January 31). Research guides: Factiva user guide: Results screen. University at Buffalo, University Libraries. Retrieved from https://research.lib.buffalo.edu/ factivaguide/results Loughran, T., & McDonald, B. (2011). When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. The Journal of Finance, 66(1), 35–65. Raimondo, C. (2019). The media and the fnancial markets: A review. Asia-Pacifc Journal of Financial Studies, 48(2), 155–184. https://doi.org/10.1111/ajfs.12250 Root, A. (2020a, August 5). Nikola stock is taking a hit after earnings. Here’s what wall street is saying. —Barrons.com. Dow Jones Institutional News. http://global.factiva.com/redir/default. aspx?P=sa&an=DJDN000020200805eg850023f&cat=a&ep=ASE Root, A. (2020b, December 23). Nikola got some very bad news. Here’s how low the stock could go. —Barrons.com. Dow Jones Institutional News. Retrieved from http://global.factiva.com/ redir/default.aspx?P=sa&an=DJDN000020201223egcn001l3&cat=a&ep=ASE Whitehouse, M. (2022). Writing in fnance: Improving the communicative potential of fnancial analysts’ recommendations [Dissertation, Università della Svizzera italiana]. Retrieved from https://digitalcollection.zhaw.ch/handle/11475/24921 Whitehouse, M., Palmieri, R., & Perrin, D. (2018). The pragmatics of fnancial communication. Part 2: From public sphere to investors. International Journal of Business Communication, 55(3), 267–274. https://doi.org/10.1177/2329488418779206 Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made. The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. 7 The Hindenburg Report as a Turning Point