1 / 21

The challenge to asses the consequences of LBO on companies “Quantitative” Evidence from Europe.

The challenge to asses the consequences of LBO on companies “Quantitative” Evidence from Europe. Nicolas BEDU nicolas.bedu@u-bordeaux4.fr Groupe de Recherche en Économie Théorique et Appliquée (GREThA) UMR CNRS 5113, Montesquieu - Bordeaux IV University.

arlo
Download Presentation

The challenge to asses the consequences of LBO on companies “Quantitative” Evidence from Europe.

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. The challenge to asses the consequences of LBO on companies “Quantitative” Evidence from Europe. Nicolas BEDU nicolas.bedu@u-bordeaux4.fr Groupe de Recherche en Économie Théorique et Appliquée (GREThA) UMR CNRS 5113, Montesquieu - Bordeaux IV University

  2. The need of large quantitative analysis. • The impact of LBO on employment, default and bankruptcy. • Discussion about methods. • Discussion about data. • The “French case”. • The interest of an European study ?

  3. The need of large quantitative analysis (1) • What are the consequences of LBO ? • defaults and bankruptcy ? • Lay offs ? • Profitability and productivity ? • Methods: • Databases studies • Most comprehensive methods to analyse global trends. • Survey studies • This method allows to capture more details than databases studies (qualitative informations). Nevertheless, it suffers from a huge survivor bias • Case studies • This method permits to capture high detailled informations (interview) but it does not allow to observe global trend

  4. The need of large quantitative analysis (2) • Sample size in recent empirical studies (consequence on employment)

  5. The need of large quantitative analysis (3) • Sample size in recent empirical studies (financial distress and bankruptcy)

  6. The need of large quantitative analysis (4) • Sample size in recent empirical studies (economic performance and productivity)

  7. The impact of LBO on employment • No consensus about the impact of LBO on employment. • We argue that the methods used to analyse the effect of LBO on employment affect the sense of the relationship • Methods based on a comparison between the firms which are involved in LBO and average of the whole sector in terms of employment (Chaplinsky et al., 1998; Kaplan, 1989b, Smith, 1990; Weir et al., 2008). These studies conclude a negative relationship between LBO financings and employment. • There is no proof of a significant impact of LBO on employment in studies dealing with endogeneity fo the LBO decisions (Amess and Wright 2007a, 2007b) • Matching methods: • Job destruction (Davis et al., 2008; Cressy et al., 2008) • Indeterminate effect on employment (Amess and Wright 2007a, 2007b; Amess et al., 2008) • Job creation (Boucly et al., 2008, 2009)

  8. The impact of LBO on default and bankruptcy of targeted companies • A consensus about financial distress, default and bankruptcy for targeted companies : LBO financing does not increase the risk of failure. • No specific risk for LBO backed companies (Boucly et al. 2009) • “PE backed buyouts post 2003 are not riskier than the non-buyout population” (Wilson et al. 2010) • A lower failure rate for targets when a PE fund is involved in the deal. (Wilson et al. 2010) • However, others interpretation could be proposed: • What is the ability of PE funds to improve the management of targeted company ?. • What is the propensity of failures after the exit of PE fund ? • Is the debate centred on the method of financing or the role of PE fund ? • What is the impact of debt leverage ? Does debt leverage increase the risk of failure ? • A specific risk for LBO financing => how to estimate the debt leverage ? • Are data about debt amounts available ? • Debt amounts are declared in the balance sheet of the holding. • Not data available on holding

  9. Matching methods (1) • “Matching” methods : the building of a counterfactual or a control sample for estimating the impact of LBO on the level of employment or failure/bankruptcy. • “Matching” methods on classic variables • Sectoral affiliation (Boucly et al., 2009; Cressy et al., 2008; Davis et al., 2008) • Size (Boucly et al., 2009; Cressy et al., 2008; Davis et al., 2008) • Performance prior LBO (Boucly et al., 2009; Cressy et al., 2008; Davis et al., 2008) • Level of employment prior LBO (Amess and Wright 2007a, 2007b) • Cost of labour per employee and profit per employee (Amess and Wright 2007a, 2007b) • “Propensity score matching” methods: two step-method • To identify the effects of observable variables on the probability for firm to undergo a LBO (Probit model). • by using the propensity score obtained through the Probit model, the impact of LBO on dependant variable could be estimated through a difference-in-differences regression

  10. Matching methods (2) • Interests of “Matching” methods • the occurrence of an event may not be randomly distributed but along some characteristics of statistical individuals. • The impact of the independant variable could capture only the heterogeneity between the two samples. • Critics : A robust counterfactual? There is a great heterogeneity of ownership status for firms involved in a LBO: listed companies, SME, subsidiaries, family business, etc. This heterogeneity involves paying attention to the different characteristics used to build the counterfactual. • Given the specific ownership status, we have to make a difference between the holding which has the property rights and the target company. • Most of the studies focus on the private equity markets in the UK or in the USA. We assume that the studies of other countries, especially European countries need to pay attention to national specificities in terms of rigidity of employment, depth of financial markets, size of firms and ownership status. • The decision of the buyout could be explained through non observable variables.

  11. How to measure the impacts of LBO on employment and failures (1) • According to the mentioned advantages and limits about “matching methods”, we give some proposal of a comprehensive “matching” method: • Using a propensity score matching is relevant in order to avoid some bias in the estimation of the impact of LBO on employment. We argue that companies which undergo a LBO present some specific characteristics. • According to Boucly et al. (2009) we have to control the probability of failure which must be similar between target and control. • A complex corporate structure could induce a bias because of restructuring like mergers between establishments. As mentioned by Davis et al. (2008), the estimation of the impact of LBO on employment needs to distinguish the regressions at firm-level and plant-level. • Macroeconomic effect has to be controlled. We assume the relevance of cyclical effect in the LBO industry. For example, one can argue that returns, risks of failures, layoff trends were quite different before the financial crisis of 2007, especially for the LBO market which was in a bubble prior to the crisis. • According to Amess and Wright (2007b), the impact of buyout could vary if the deal implies internal managers (MBO) or external managers (MBI).

  12. How to measure the impacts of LBO on employment and failures? (2) • One can argue that buyout sample are quite different and studies focus on different countries: the USA, the UK and France. Most of the studies on the UK give evidence about an insignificant role of LBO in a fall in employment. Davis et al. (2008) find that LBO has a negative impact on the level of employment for the USA. On the contrary, Boucly et al. (2008; 2009) highlight that LBO plays a significant role in job creation. Different findings imply to consider some national specificities for explaining the role of LBO on employment. • Given the great heterogeneity of ownership status of companies prior to LBO, control samples must be built without any bias towards a specific category of ownership status. In fact, it would be easier to obtain information for companies which are listed, but, buyouts involve more often, private companies • Control sample has to deal with the maturity of national private equity industry. We assume that there are potentially more highly profitable deals in a less mature private equity market. • Given the debates about the role of LBO funds and the heterogeneity of the operating performance between targets, we propose the need to distinguish companies along their economic and financial health. The relationship between LBO and employment could be rather different if we deal with distressed firms or “cash cows”.

  13. Discussion about data • Databases on deals and databases on companies • Databases on deals: • Thomson Reuters • Zephyr (Bureau Van Dijk) • Databases on targeted companies • Thomson Reuters • Amadeus (Europe) • Diane (France) • Coverage and quality of data • Longitudinal monitoring available? • Databases are affected by a survivor bias and mostly, do not provide enough information about relevant characteristics to build a control sample. • The need of hand-made data set (data about employment in Amadeus an Diane database)

  14. Coverage of data (1999-2008) (1)

  15. Coverage of data (1999-2008) (2)

  16. Coverage of data (1999-2008) (3)

  17. Coverage of data (1999-2008) (4)

  18. Coverage of data (1999-2008) (5)

  19. Coverage of data (1999-2008) (6)

  20. French administrative data. • Given the limits of Amadeus and Thomson Reuters databases, we argue the interests of “administrative” data produced by INSEE (National Institute of Statistics and Economic Studies): • “objective” data • Exhaustive data • Longitudinal monitoring available • DADS files (INSEE) : Annual Declaration of Social Data. • Individual data on employed workforce for all French companies. • Matching on SIREN (National Enterprise and Establishment Register Database) • Free of charge… but data access is restricted (statistical secret) • Bankruptcy file (Altares-INSEE). • Data on receivership and compulsory liquidation. • Matching on SIREN.

  21. Analysing the consequences of LBO on companies : a Pan-European Study ? • ”Home bias” of researchers : • National data are difficult to find for foreign researchers (difficulty to obtain an access to French administrative data) • Specificity of national legislation concerning notification of employment, declaration of bankruptcy, etc. • “Home bias” of empirical studies : UK and France. • Specificity and configurations of national LBO markets (size and maturity, legal and fiscal environment for Private Equity, degree of financialization, regulation of labour markets, etc.) • A “still” actual debate… with very few large-scale empirical studies

More Related