Skip to main content

The impact of investor sentiment on the German stock market


This paper develops a broad-based sentiment indicator for Germany and investigates whether investor sentiment can explain stock returns on the German stock market. Based on a principal component analysis, we construct a sentiment indicator that condenses information of several well-known sentiment proxies. We show that this indicator explains the return spread between sentiment sensitive stocks and stocks that are not sensitive to sentiment fluctuations. Specifically, stocks that are difficult to arbitrage and hard to value are sensitive to the indicator. However, we do not find much predictive power of sentiment for future stock returns.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2


  1. 1.

    We use six-months expectations instead of one-month expectations because Sentix one-month horizon answers are very noisy (Schmeling2007).

  2. 2.

    As an alternative test of the usability of the Sentix and the G-Mind as sentiment indicators, we also conducted the same analysis as we did to validate our sentiment indicator GSI: We relate both alternative sentiment indicators, G-Mind and Sentix, to the differences between contemporaneous returns of stocks that are sensitive to sentiment fluctuations and those that are not. Our results (not reported) show that G-Mind(Stocks) as well as Sentix generally have no significant impact on the contemporaneous return spread between sentiment sensitive and sentiment insensitive stocks.

  3. 3.

    Data on trading volume is provided by Deutsche Bundesbank, the total number of listed firms are collected from DeutschesAktieninstitut’s (DAI) Factbooks. The DAI Factbook is updated once a year and contains a comprehensive collection of statistics on the German stock market.We trend adjust trading volume as in Andersen (1996) and compute

    $${\text {TradVol}}_{\text {m}}^{\text{ta}} = {\text{LN}}\left( {\frac{{{\text {TradVol}}_{\text{m}}}}{{{\text {TotalFirms}}_{\text{m}}}}} \right) - \frac{1}{{24}}\sum\nolimits_{{\text{m}} = - 24}^{{\text{m}} = - 1} {\text{LN}} \left( {\frac{{{\text {TradVol}}_{\text{m}}}}{{{\text {TotalFirms}}_{\text{m}}}}} \right)$$

    whereTradVol m is trading volume in million EUR in monthm, andTotalFirms m is the total number of listed firms in monthm.

  4. 4.

    Another proxy regularly used in U.S. studies is the closed end fund discount (see, e.g., Lee et al. (1991) and Neal and Wheatley (1998)). Since listed closed end equity funds do not exist in Germany, we cannot use this proxy in our study. However, net flows are the mutual fund equivalent of the closed end fund discount. Thus, we think that we do not miss an important aspect of sentiment by not being able to include the closed end fund discount.

  5. 5.

    Using value-weighted IPO returns instead does not affect our results (not reported).

  6. 6.

    The Hoppenstedt Aktienführer is updated annually and contains information about all listed German firms including balance sheet and profit and loss items.

  7. 7.

    The resulting first principal component is also normalized and has a mean of zero and a standard deviation of one.

  8. 8.

    The only exception is the impact of trading volume, which does not play a large role in the sub-period 2000–2006 anymore. However, this does not affect our results: The correlation between our GSI and the first component of a PCA estimated based on the same proxies as above but excluding trading volume is 99.98% in the 2000–2006 period.

  9. 9.

    The time series data for the GSI is provided for the use of other researchers under

  10. 10.

    We follow Baker and Wurgler (2006) and use equal weighted portfolios, because large firms will probably be less affected by sentiment. Thus, using value weighting will “tend to obscure the relevant patterns”. (p. 1646)

  11. 11.

    For the sake of brevity, we use GSI as an abbreviation for the German Sentiment indicator in the main text. It comprises both, the macro-adjusted indicator\(\left( {\textit{GSI}}_{m}^{A} \right)\) and the unadjusted indicator\(\left( {\textit{GSI}}_{m} \right)\).

  12. 12.

    The list of sorting criteria to define sentiment sensitive stocks is long. We mainly follow the article by Baker and Wurgler (2006) as a benchmark for the choice of sorting criteria. However, they also include additional proxies like expenditures for R&D or sales growth for which we could not get data for all of the firms in our sample.

  13. 13.

    Another proxy for limits of arbitrage is stock illiquidity (Kumar and Lee (2006)). Thus, illiquid stocks might be particularly prone to sentiment fluctuations, too. However, at the same time Baker and Stein (2004) argue that high liquidity is a proxy for investor sentiment because it signals that the market is currently dominated by irrational investors. Thus, overall it is not clear how liquidity is related to sentiment. In unreported tests using the Amihud (2002) illiquidity ratio as a liquidity proxy we find no clear relationship between liquidity levels and sentiment fluctuations.

  14. 14.

    DAFOX and CDAX have very similar return characteristics. In the overlapping period from January 1993 to December 2004 the monthly returns of DAFOX and CDAX are almostperfectly correlated (correlation coefficient of 0.978). The correlation coefficient between the monthly returns of the combined DAFOX/CDAX and a value-weighted index based on all stocks considered in our study is 0.979. Thus, the DAFOX/CDAX serves as an appropriate market proxy for our stock universe.

  15. 15.

    SMB m andHML m are constructed for the German stock market as described in Artmann et al. (2011), while the design ofWML m generally follows Carhart (1997).

  16. 16.

    Interestingly, the non-sentiment sensitive stock portfolio Q(5) delivers a positive abnormal return, as indicated by the significantly positive intercept presented in the second to last column ofTable 6, while the sentiment sensitive stock portfolio does not. This suggests that sentiment sensitivity is not a positively priced risk factor on the German stock market. This result is also consistent with recent findings in Koch (2010), who finds that German stocks with high idiosyncratic risk earn negative abnormal returns.

  17. 17.

    Results for a not macro-adjusted sentiment indicator (not reported) are very similar.

  18. 18.

    Alternatively, we compute idiosyncratic volatility relative to a one-factor model. Results (not reported) are virtually unchanged.

  19. 19.

    For supportive evidence, see also Lee and Radhakrishna (2000). However, note that Hvidkjaer (2006) also shows that institutional investors make smaller trades in small stocks. Thus, trade size might also partially proxy for firm size. We examine the role of firm size explicitly in Section 4.2.

  20. 20.

    Alternatively, we conduct this analysis for the first ten years of our sample period (19932003) only to make sure that our results are not biased by gradual trading of institutional investors that might have already slowly started in the last years of our sample. Results (not reported) are stable.

  21. 21.

    Data on firm foundation dates are hand-collected from Hoppenstedt Aktienführer. The results do not change, if we measure firm age in number of month since the firm’s first appearance in our sample.

  22. 22.

    Market value of equity is computed as stock price times shares outstanding. Data on shares outstanding are from Hoppenstedt Aktienführer.

  23. 23.

    Alternatively, we sort firms into profitability quintiles. Our results (not reported) remain stable.

  24. 24.

    In unreported tests, we find no strong effects if we relate contemporaneous sentiment to future aggregate returns.

  25. 25.

    In unreported tests, we also examine longer periods like months 13 to 24. We generally find no significant results for these longer horizons.

  26. 26.

    As sentiment is standardized, this means that\({\textit{GSIDUM}}\left( {\textit{Lo}} \right)_m^A.{\textit{GSIDUM}}\left( {\textit{Hi}} \right)_m^A\left( {{\textit{GSIDUM}}\left( {\textit{Lo}} \right)_m^A} \right)\) is one if sentiment in monthm is more than one standard deviation above (below) its mean.

  27. 27.

    Unfortunately, Baker et al. (2009) do not provide country-level results.

  28. 28.

    Note, that the sentiment indicator analyzed so far is calculated ex post. In unreported analysis, we also examine an alternative indicator which is again based on PCA. However, in contrast to the proxy used above, it is calculated at each point in time only using a rolling window of past data. This indicator is highly correlated with the aggregate sentiment measure used before and we find very similar results: Again, there is no strong predictive power of sentiment for future aggregate returns or return spreads.

  29. 29.

    Studies that find signs of irrational behavior among professional investors include Haigh and List (2005), Coval and Shumway (2005), Glaser et al. (2010), and Puetz and Ruenzi (2011).


  1. Amihud Y (2002) Illiquidity and stock returns: cross-section and time-series effects. J Financ Mark 5:31–56

    Article  Google Scholar 

  2. Andersen TG (1996) Return volatility and trading volume: an information flow interpretation of stochastic volatility. J Financ 51:169–204

    Article  Google Scholar 

  3. Andres C (2008) Large shareholders and firm performance—an empirical examination of founding-family ownership. J Corp Financ 14:431–445

    Article  Google Scholar 

  4. Artmann S, Finter P, Kempf A, Koch S, Theissen E (2011) The Cross-Section of German Stock Returns: New Data and New Evidence. Schmalenbach Bus Rev (Forthcoming)

  5. Baker M, Stein J (2004) Market liquidity as a sentiment indicator. J Financ Mark 7:271–299

    Article  Google Scholar 

  6. Baker M, Wurgler J (2000) The equity share in new issues and aggregate stock returns. J Financ 55:2219–2257

    Article  Google Scholar 

  7. Baker M, Wurgler J (2006) Investor sentiment and the cross-section of stock returns. J Financ 61:1645–1680

    Article  Google Scholar 

  8. Baker M, Wurgler J (2007) Investor sentiment in the stock market. J Econ Perspect 21:129–151

    Article  Google Scholar 

  9. Baker M, Wurgler J, Yuan Y (2011) Global, local, and contagious investor sentiment. J Financ Econ (Forthcoming)

  10. Barberis N, Shleifer A, Vishny R (1998) A model of investor sentiment. J Financ Econ 49:307–343

    Article  Google Scholar 

  11. Betrand M, Mullainathan S (2001) Do people mean what they say? Implications for subjective survey data. Amer Econ Rev 91:67–72

    Article  Google Scholar 

  12. Bram J, Ludvigson SC (1998) Does consumer confidence forecast household expenditure? A sentiment index horse race. Federal Reserve Bank of New York. Economic Policy Review 4:59–78

    Google Scholar 

  13. Brown GW, Cliff MT (2004) Investor sentiment and the near-term stock market. J Empirical Finance 11:1–27

    Article  Google Scholar 

  14. Brown GW, Cliff MT (2005) Investor sentiment and asset valuation. J Bus 78:405–440

    Article  Google Scholar 

  15. Burghardt M, Czink M, Riordan R (2008) Retail investor sentiment and the stock market. Working Paper

  16. Campbell JY (2003) Discussion of “perspectives on behavioral finance: does irrationality disappear with wealth?” NBER Macroeconomics Annual

  17. Carhart MM (1997) On persistence in mutual fund performance. J Financ 52:57–82

    Article  Google Scholar 

  18. Conrad J, Cornell C, Landsman WR (2002) When is bad news really bad news? J Financ 57:2507–2532

    Article  Google Scholar 

  19. Cornelli F, Goldreich D, Ljungqvist A (2006) Investor sentiment and pre-IPO markets. J Financ 61:1187–1216

    Article  Google Scholar 

  20. Coval J, Shumway T (2005) Do behavioral biases affect prices? J Financ 60:1–34

    Article  Google Scholar 

  21. Dennis P, Mayhew S (2002) Risk-neutral skewness: evidence from stock options. J Financ Quant Anal 37:471–493

    Article  Google Scholar 

  22. Derrien F, Kecskes A (2009) How much does investor sentiment really matter for equity issuance? Europ Finan Manage 15:787–812.

    Article  Google Scholar 

  23. Dorn D (2009) Does sentiment drive the retail demand for IPOs? J Financ Quant Anal 44:85–108

    Article  Google Scholar 

  24. Faccio M, Lang LHP (2002) The ultimate ownership of western European corporations. J Financ Econ 65:365–395

    Article  Google Scholar 

  25. Fama EF, French KR (1993) Common risk factors in the return on bonds and stocks. J Financ Econ 33:3–53

    Article  Google Scholar 

  26. Fisher K, Statman M (2000) Investor sentiment and stock returns. Financ Anal J March/April:16–23

  27. Frazzini A, Lamont OA (2008) Dumb money: mutual fund flows and the cross-section of stock returns. J Financ Econ 88:299–322

    Article  Google Scholar 

  28. Gao X, Yu J, Yuan Yu (2010) Investor sentiment and idiosyncratic risk puzzle. Working Paper

  29. Glaser M, Langer T, Weber M (2010) Overconfidence of professionals and lay people: individual differences within and between tasks? Working Paper

  30. Glushkov D (2009) Sentiment beta. Working Paper

  31. Groves RM (2006) Nonresponse rates and nonresponse bias in household surveys. Pub Opin Q 70:646–675

    Article  Google Scholar 

  32. Haigh M, List J (2005) Do professional traders exhibit myopic loss aversion—an experimental analysis. J Financ 60:523–534

    Article  Google Scholar 

  33. Hendershott T, Riordan R (2011) Algorithmic trading and information. Working Paper

  34. Hengelbrock J, Theissen E, Westheide C (2010) Market response to investor sentiment. Working Paper

  35. Hvidkjaer S (2006) A trade based analysis of momentum. Rev Finan Stud 19:457–491

    Article  Google Scholar 

  36. Hvidkjaer S (2008) Small trades and the cross-section of stock returns. Rev Finan Stud 21:1123–1151

    Article  Google Scholar 

  37. Indro D (2004) Does mutual fund flow reflect investor sentiment? J Behav Financ 5:105–115

    Article  Google Scholar 

  38. Jegadeesh N, Titman S (1993) Returns to buying winners and selling losers: implications for stock market efficiency. J Financ 48:65–91

    Article  Google Scholar 

  39. Jones C (2002) A century of stock market liquidity and trading costs. Working Paper

  40. Koch S (2010) Low risk and high returns: evidence from the German stock market. Working Paper

  41. Kumar A, Lee CMC (2006) Retail investor sentiment and return comovements. J Financ 61:2451–2485

    Article  Google Scholar 

  42. Lee CMC, Radhakrishna B (2000) Inferring investor behavior: evidence from TORQ data. J Financ Mark 3:83–111

    Article  Google Scholar 

  43. Lee CMC, Swaminathan B (2000) Price momentum and trading volume. J Financ 55:2017–2069

    Article  Google Scholar 

  44. Lee CMC, Thaler RH, Shleifer A (1991) Investor sentiment and the closed-end fund puzzle. J Financ 46:75–109

    Article  Google Scholar 

  45. Loughran T, Ritter JR (1995) The new issues puzzle. J Financ 50:23–51

    Article  Google Scholar 

  46. Neal R, Wheatley SM (1998) Do measures of investor sentiment predict returns? J Financ Quant Anal 33:523–547

    Article  Google Scholar 

  47. Newey WK, West KD (1987) A simple, positive semi-definite, heteroskedasticity and autocorrelation consistent covariance matrix. Econometrica 55:703–708

    Article  Google Scholar 

  48. Oehler A, Rummer M, Smith PN (2005) IPO pricing and the relative importance of investor sentiment—evidencefrom Germany. Working Paper

  49. Otoo MW (1999) Consumer sentiment and the stock market. Working Paper

  50. Puetz A, Ruenzi S (2011) Overconfidence among professional investors: evidence from mutual fund managers. J Bus Financ Account 38:684–712

    Article  Google Scholar 

  51. Qiu L, Welch I (2006) Investment sentiment measures. Working Paper

  52. Schiereck D, DeBondt W, Weber M (1999) Contrarian and momentum strategies in Germany. Financial Analysts Journal 55:104–114

    Article  Google Scholar 

  53. Schmeling M (2007) Institutional and individual sentiment: smart money and noise trader risk? Int J Forecasting 23:127–145

    Article  Google Scholar 

  54. Schmeling M (2009) Investor sentiment and stock returns: some international evidence. J Empirical Finance 16:394–408

    Article  Google Scholar 

  55. Schmitz P, Glaser M, Weber M (2009) Individual investor sentiment and stock returns—what do we learn from warrant traders? Working Paper

  56. Shleifer A, Vishny RW (1997) The limits of arbitrage. Journal of Finance 52:35–55

    Article  Google Scholar 

  57. Wurgler J, Zhuravskaya E (2002) Does arbitrage flatten demand curves for stocks? J Bus 75:583–609

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Philipp Finter.

Appendix: List of Variables

Appendix: List of Variables

Abbreviation Variable Source Frequency Measurement
Panel A: Sentiment Proxies
GFKm GFK Consumer Confidence Survey Bloomberg monthly Index
IPCRm Inverted Put-Call Ratio Deutsche Börse AG monthly Percent, number of calls traded divided by number of puts traded.
TradVolta m Trading Volume Deutsche Bundesbank monthly Million EUR divided by number of firms; trend adjusted as in Andersen (1996).
Flowsm Net Fund Flows Deutsche Bundesbank monthly Level
IPO-Numm Number of IPOs Deutsche Börse AG, DAI Factbooks,
Hoppenstedt Aktienführer
monthly Level
IPO-Retm IPO Returns Deutsche Börse AG, DAI Factbooks,
Hoppenstedt Aktienführer
monthly Percent
E/D-Ratiom Equity/Debt Ratio Deutsche Bundesbank monthly Level
Panel B: Macroeconomic Variables
IndProdm+x1 Industrial Production Deutsche Bundesbank monthly All macroeconomic variables are computed as month-over-month changes based on 12-month-moving-averages of the underlying index variables.
InvOrdm+x2 Inventory Orders Deutsche Bundesbank monthly
FacOrdm+x3 Factory Orders Deutsche Bundesbank monthly
RetSalm+x4 Retail Sales Deutsche Bundesbank monthly
Emplm+x5 Employment Deutsche Bundesbank monthly
Panel C: Risk Factors
RMRFm Excess Market Return Karlsruher Kapitalmarktdatenbank, Hoppenstedt Aktienführer monthly Excess market return over risk free rate.
SMBm Small Minus Big Karlsruher Kapitalmarktdatenbank, Hoppenstedt Aktienführer monthly Return difference between portfolios of small and large firms.
HMLm High Minus Low Karlsruher Kapitalmarktdatenbank, Hoppenstedt Aktienführer monthly Return difference between portfolios of high and low book-to-market equity firms.
WMLm Winner Minus Loser Karlsruher Kapitalmarktdatenbank, Hoppenstedt Aktienführer monthly Return difference between portfolios of high and low return momentum firms.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Finter, P., Niessen-Ruenzi, A. & Ruenzi, S. The impact of investor sentiment on the German stock market. Z Betriebswirtsch 82, 133–163 (2012).

Download citation


  • Investor sentiment
  • Stock returns
  • German stock market


  • G12
  • G14