Accounting information in innovative small cap firms: evidence from London’s Alternative Investment Market

Abstract

We posit that investors and social media users place more weight on cash flows than on earnings for innovative small cap firms and that, in turn, innovative small cap firms (i) manage cash flows more than earnings, and (ii) disclose more cash flow than earnings information on social media. Using a matched sample of innovative and non-innovative small cap firms listed on the London’s Alternative Investment Market (AIM), we document that the value relevance of cash flows (earnings) is higher (lower) for innovative compared to non-innovative small cap firms. Using Twitter to examine the demand of accounting performance measures, we find that Twitter users more frequently retweet and include as ‘Favorite’ information about cash flows, than information about earnings for innovative small cap firms. We then show that innovative small cap firms engage less intensively in earnings management and exhibit higher abnormal cash flows compared to non-innovative small cap firms. Innovative small cap firms emphasise more information in their tweets about cash flows and less about earnings compared to non-innovative small cap firms. Cross-sectional tests demonstrate that seasoned equity offerings provide additional incentives to engage in increasing abnormal cash flow management activities.

Keywords:

• financial reporting
• innovation
• small cap firms
• social media
• cash flows
• earnings
• AIM London

Acknowledgements

This paper is based on the first chapter of Alessandro’s dissertation at ESSEC Business School. We gratefully acknowledge the helpful comments of Paul André, Daniel Beneish, Elodie Behnam (discussant), Anastasios Elemes, Beatriz Garcia Osma, Martin Glaum (discussant), Thomas Jeanjean, Katherine Schipper, and Teri Yohn and of workshop participants at the 2016 BPF Campus, 2016 SIDREA Conference, 2017 EAA Conference, 2017 AFC Conference, and 2018 IAAER World Congress.

Disclosure statement

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

Notes

1 The decision by Bloomberg Database, which provides current and historical financial quotes, and business newswires on over 52,000 companies worldwide, to include real-time tweets as part of its news feed demonstrates that investors value corporate social media information. See: https://www.bloomberg.com/professional/ (accessed 3 August 2020).

2 Twitter changed the name from ‘Favorite’ to ‘Like’ in November, 2015 to indicate an appreciation of a certain tweet (Source: https://blog.twitter.com/official/en_us/a/2015/hearts-on-twitter.html).

3 For further details on the value relevance of accounting performance measures across firms with different characteristics, please also see: Barth et al. (2001), Holthausen and Watts (2001), Kallapur and Kwan (2004), Landsman (2007), Wyatt (2008), Song et al. (2010).

4 https://www.londonstockexchange.com/companies-and-advisors/aim/aim/aim.htm.

5 http://www.londonstockexchange.com/statistics/historic/aim-country-of-operation-and-incorporation/aim-companies-country-of-operation.htm.

6 To mitigate concerns that limiting our sample to AIM London listed firms may result in a small sample size, we compared the size of the sample with qualifying firms listed on NASDAQ. Using the same cut-off (€43 million total assets), we obtain a sample of similar size. NASDAQ-listed firms are small, but on average larger than AIM London ones.

7 As the main definition of small and medium sized enterprises follows the EU recommendation 2003/61, all data used and reported in the paper is expressed in Euros.

8 All results reported in this study are qualitatively similar if we use the Industry Classification Benchmark (ICB) as the industry classification (Achleitner et al. 2014).

9 We set to zero any missing R&D expenses (Barth et al. 2018). This approach is consistent with the materiality principle in accounting standards applied by firms listed on AIM. Firms would not need to disclose R&D expenditures when this accounting number is not material.

10 Results are similar if we use other winsorization levels (1% or 5%).

11 Gerakos et al. (2013) report that between 1995 and 2008, the average (median; 25th percentile) 12-month return for their sample firms listed on AIM London is −13% (−18.5%; −55.2%). To reconcile the different results, we compute returns only for large firms, and not at small cap firms, and for the period prior to 2008. We obtain results more aligned to Gerakos et al. (2013) (median: −6.68%; mean: −0.87%; p25: −45%). In addition, Gerakos et al. (2013) focus on the post-listing period (12-, 18-, 24-months post IPO) when returns might be more volatile.

12 Earnings are calculated as earnings before extraordinary items, and cash flows (CFO) are cash flows from operations as reported in the statement of cash flows. Consistent with Bushman et al. (2016), we estimate total accruals (TACC) from the statement of cash flows to avoid the well-documented measurement errors of balance sheet-based accruals. We also re-perform our tests by using a balance sheet approach to estimate total accruals and cash flows. Results remain very similar.

13 The inclusion of firm fixed-effects in this setting is not a good model specification for our sample due to the lack of within-firm variation in our sample. Equations including firm fixed-effects allow researchers to investigate within-firm variations in the dependent and independent variables. However, R&D expenses tend to be persistent over time – firms tend to commit to certain R&D expenditure levels because, on average, innovation is a long-term investment. Thus, innovative small cap firms tend to stay as such over the period and/or grow, moving out of our cut off (i.e. above €43 million total assets) and/or move onto the main London Stock Exchange market. Specifically, our sample includes 148 innovative small cap firms, 152 non-innovative small cap firms – only 21.3% of firms change category over the period analysed.

14 We compute the odds ratio as e^ß= e^0.087 = 1.10 (Miller and Rodgers 2008).

15 We compute the odds ratio as e^ß= e^0.316 = 1.37 (Miller and Rodgers 2008).

16 We re-performed equation (2) by using continuous social media variables. The results are qualitatively similar. The only exception is the comparison between FAV_EARN and FAV_CASH where the difference in the coefficients of interest is positive, but not statistically different.

17 A potential endogeneity concern is that the market measure, market-to-book ratio, drives our findings about managerial decisions. Our results are substantially identical when we delete market-to-book ratio as a control variable.

18 We compute the odds ratio as e^ß= e^–0.372 = 0.69 (Miller and Rodgers 2008).

19 We re-performed equation (5) by using continuous variables to measure corporate social media activity. The results are virtually identical. On average, innovative small cap firms tweet 54% more tweets about cash flows than non-innovative small cap firms. Innovative small cap firms also use on average 26% more hashtags associated with cash flow information than non-innovative small cap firms.

20 https://www.raconteur.net/sponsored/uk-businesses-must-adapt-catch-ip.