Table of Contents
Table of Contents
List of Tables
List of Abbreviations
2. Characteristics of Retail Private Equity
2.1 Determinants of the Industry Environment
2.2 Private Equity Value Creation in Retail
3. Literature Review and Development of Hypotheses
3.1 General Partner Experience
3.2 Private Equity Fund Size
3.3 Retail Industry Specialization
4. Data and Methodology
4.1 Data Compilation and Restrictions Applied
4.2 Data Quality
4.3 Descriptive Statistics
4.5.2 Independent Variables
4.5.3 Control Variables
5. Empirical Results
5.1 Performed Regressions
5.2 Tests of Model Specifications
5.4 Discussion of Limitations and Scope for Future Research
This thesis examines the relationship between private equity firm and fund level characteristics and the performance of leveraged buyouts in the retail sector. More specifically, the three dimensions general partner experience, fund size, and the fund’s retail specialization are analyzed. The impact of the three characteristics on the investment performance proxied by money multiples and internal rates of return, as well as by the write-off probability of the buyout, is assessed by analyzing a sample of 10,376 buyouts conducted since 1993, of which 614 deals are transactions in the retail sector. The findings indicate that the experience of a private equity firm is no suitable indicator for superior investment performances anymore. Moreover, a private equity fund’s size is found to negatively impact its returns and at the same time, increases the write-off probabilities of the fund’s investment. Retail companies bought by large funds tend to underperform even stronger than the other investments conducted by the same vehicle. Finally, funds specialized in retail do achieve higher returns in their retail transactions, while simultaneously facing a higher probability of those retail investments resulting in bankruptcy.
List of Exhibits
Exhibit 1: Number of retail LBOs, LBO write-offs in retail, overall dry powder (2000-2018)
Exhibit A 1: MSCI World Large- and Mid-Cap Index development (Jan. 1993 to Apr. 2019)
Exhibit A 2: Yearly no. of funds raised and global buyout fundraising volumes (1992 to2018)
List of Tables
Table 1: Multivariate Linear Regressions ofMoney Multiple, Sample I
Table 2: Multivariate Linear Regressions ofMoney Multiple, Sample I
Table 3: Logistic Regressions ofWrite-offProbability, Sample Ia
Table 4: Logistic Regressions ofWrite-offProbability, Sample I
Table A 1: Number of Investment Exits perYear, Sample I
Table A 2: Number of Investment Exits perYear, Sample Ia
Table A 3: Summary Statistics on Performance
Table A 4: Summary Statistics Fund Sequence per Deal
Table A 5: Summary Statistics Fund Size in USD MM, Deal Level
Table A 6: Summary Statistics Fund Fraction ofRetail Investments, Deal Level
Table A 7: Summary Statistics Holding Period, Deal Level
Table A 8: Summary Statistics Total Deals per Invested Fund, Deal Level
Table A 9: Summary Statistics Buyouts by Geography, Deal Level
Table A 10: Summary Statistics Buyouts by Sector, Deal Level
Table All: Variance Inflation Factors for Money Multiples, Sample I
Table A 12: Variance Inflation Factors for IRR, Sample I
Table A 13: Variance Inflation Factors forWrite-offs, Sample I
Table A 14: Multivariate Linear Regressions ofIRR, Sample Ia
Table A 15: Multivariate Linear Regressions ofIRR, Sample I
Table A 16: Multivariate Linear Regressions ofMoney Multiple, S. I, excl. Retail Deals
Table A 17: Multivariate Linear Regressions ofIRR, S. I, excl. Retail Deals
Table A 18: Logistic Regressions ofWrite-offProbability, S. I, excl. Retail Deals
List of Abbreviations
Abbildung in dieser Leseprobe nicht enthalten
According to Bloomberg News, the year of 2017 was known as the “Retail Apocalypse” with more U.S. retailers filing for bankruptcy than during the year of the financial crisis (Townsend, Surane, Orr & Cannon, 2017). By taking a closer look at the reasons for these defaults, one will quickly understand that the culprit might not solely be Amazon Inc. tapping on market share. Before their bankruptcies, numerous retail companies were bought and highly levered by private equity firms who neither were able to implement adequate operational improvements, nor product and sales innovations. The resulting collapses like the one of toy giant Toys”R”Us in September 2017, twelve years after its leveraged buyout by the mega funds of Bain Capital and KKR, lead to numerous store closings andjob losses (DiNapoli & Rucinski, 2017). Hence, there is certainly a need to understand the factors determining the success of private equity investments in the retail industry. This thesis aspires to establish a causal relationship between macroeconomic, firm, and fund level characteristics and buyout performance in the retail sector.
Differences in GP experience, fund size, and the GP’s retail industry focus are hypothesized to explain investment performance in terms of money multiples and IRR, as well as investment failure proxied by its write-off probability, like in the case of Toys”R”Us. Implications from preceding literature on private equity performance will be utilized in order to derive the hypothesis. Regarding the financing stage, I will focus on buyout investments and do not include venture and growth capital deals, nor restructurings and turnarounds, as they are subject to different dynamics. The performed analysis is based on a data set of 2,325 funds and a sample of 614 individual retail transactions, derived by Preqin. The results will be compared to the findings from the full data sample consisting of 10,376 buyouts across all sectors, in orderto evaluate whether buyouts in the retail sector are subject to unique success factors.
In order to examine the hypothesis, I perform multivariate linear and logistic OLS regressions with MoM, IRR, and the investment write-off as depended variables. Additionally, I control for the investments holding period, the return to the MSCI world index in the buyout year, the total buyout funds raised in the year preceding the investment entry as well as for time and region fixed effects. While GP experience appears to have no significant effects on investment performance, fund size is found to negatively impact returns and at the same time, increases write-off probabilities. Buyouts in the retail sector conducted by large funds tend to underperform even stronger than the ones in other industries. Funds specialized in retail do achieve greater returns in retail transactions, but their retail investments are at the same time subject to heightened probabilities of the bankruptcy. As the results observed for the cross-sectoral, full data sample yield different coefficients and significances compared to the regressions on the smaller retail sample, it is concluded that investments in the retail sector are subject to dynamics and success factors that differ to the ones in other industries.
It was not until the 1980ies when the first large wave of private equity LBO’s emerged (Kaplan, 1989, p. 218).1 With its deal flows increasing, the PE sector became an advancing target of academic research.2 Thereby, an extensive body of literature on buyouts was created, focusing predominantly on their causes, as well as on their consequences on a macroeconomic and target company specific level (see i.a. Jensen, 1986; Jensen, 1988, Kaplan, 1989). Only gradually an additional strand of research started to examine PE performance in terms of returns. Initially, this was focusing on the relative performance of funds compared to public markets (i.a. Kaplan & Schoar, 2005, p. 1791).3 Only in recent years, academics started evaluating the determinants of LBOs returns, first on fund and later on deal levels, where this paper aims to make a contribution.
Still, findings on success factors on a deal specific level are found to be scarce (Aigner et al., 2008, p. 63). Only four studies have been identified, conducting empirical analysis on the influence of macroeconomic, fund and investment level characteristics on the performance of individual buyouts, yet with different motives. While Lopez-de-Silanes, Phalippou and Gottschalg (2015, p. 377) examine potential diseconomies to scale faced by PE funds, Braun, Jenkinson and Stoff (2017, p. 1) analyze whether performance is persistent among the same GP. Only Achleitner, Braun and Engel (2011, p. 147) and Thaker (2014, p. 5) allocate the main focus of their research to establishing a causal relationship between GP skills and buyout returns. Albeit they succeed to do so, their research leaves extensive scope for further analysis. Firstly, the analyzed datasets go no further than 2010 and 2012 respectively, and thereby cannot account for recent industry developments. Secondly, both papers do not distinguish between the individual industries the investments are made in, but rather analyze the private equity sector as whole. Since different sectors like healthcare, TMT and retail are often subject to highly specific dynamics, a distinct analysis of the individual value drivers and success factors is of interest for GPs, but also LPs and the target companies themselves.
In addition to money multiples and IRR, this thesis also analysis the determinants of write-offs of retail investments, as a measure of investment failure. The write-off has attracted very little theoretical attention in private equity literature. Previous deal level studies started with examining the characteristics of the more frequently chosen exits channels like IPOs, trade sales and secondary buyouts (see i.a Achleitner, Bauer, Figge and Lutz (2012, p. 2), Achleitner and Figge (2014, p. 407) and Schmidt, Steffen and Szabo (2010, p. 24)).4 This is the reason for findings on the determinants of investment failure still being scarce. Only Schmidt et al. (2010, p. 24) include the write-offs probabilities of buyouts into their research but focus their analysis on the influence of macroeconomic variables and the investments holding duration on failure, instead of the relationship between fund level characteristic and write-off probabilities.
In light of the fragmented and partly contradictory research on private equity returns and investment failure, as well as due to the little attempt that has been made to examine buyout sector sub-categories, like the retail industry, this thesis’ scope and focus are novel. Retail investments are not considered and studied as a byproduct of an overall performance analysis of all buyout industries but regarded as a specific sector experiencing distinct value drivers. Further, the available investment level data allows analyzing individual LBOs, first in in terms of money multiple and IRR performance, and second in regard to exit channels. Consequently, this paper contributes to research with a study on GP experience, fund size and fund focus - and their effects on the performance of retail private equity deals.
This thesis is structured as follows: The second chapter provides a comprehensive overview of the retail private equity market and examines the specificities of value creation in retail investments. In Chapter 3 the relevant literature is discussed and synthesized with the knowledge about retail investments to derive the research hypotheses. These hypotheses are tested in the empirical analysis: Chapter 4 describes the selected data set and its characteristics. Furthermore, it outlines the regression models and introduces the variables therein. Chapter 5 provides and discusses the regression results, examines their robustness and thereby gives initiations for further research, before Chapter 6 concludes.
2. Characteristics of Retail Private Equity
This chapter aims to provide a comprehensive overview of the global retail industry with respect to private equity investments. Firstly, industry characteristics and trends are outlined to develop a clear understanding of the key challenges GPs investing in retail face. In the second part, the activities ofPE firms in the sector are examined, and specific value creation levers, as well as their potential drawback, are assessed.
2.1 Determinants of the Industry Environment
According to eMarketer, worldwide retail sales in the year of 2018 are proxied to amount USD24,86 trillion, turning the retail industry into one of the largest worldwide in terms of revenue (eMarketer, 2017). Despite mid-range single-digit global sales growth forecasts, “the retail sector in most of the world is struggling” (Masters, 2019). Underlying causes for this include rapid changes in customer preferences as well as the fundamental disruption of the industry’s competitive environment. For the subsequent analysis, the following three developments are of particular relevance.
Digital-Centric Customer Relationships
Albeit the fact that in 2018, 90 percent of worldwide retail sales were yet generated in physical “bricks and mortar” stores, digital customer interactions became inevitable forthe sector (Deloitte, 2018, p. 6). This stems from today’s consumers being “channel agnostic” in their purchasing process, pressuring retailers into providing a frictionless customer experience across all on- and offline channels.5 To implement such omni-channel approaches, companies are required to develop sophisticated, holistic marketing and sales strategies, regardless of whether the actual purchase is generated on- or offline (McKinsey, 2018, p. 6). In order to implement those, vast investments in technological capabilities, including customer analytics and back-end software development are required (Gray, 2019).
The Competition ofDisruptors
The rise of online, as well as multi- and omni-channel strategies has paved the way for disruptive, upstart retailers who can effectively poach the already raddled customer loyalty to established players (McKinsey, 2018, p. 5).6 The success of Amazon Inc. is a paradigm for revolutionizing a long- established industry. While traditionally, a separate bricks and mortar store had to be visited for each product category, Amazon created a digital marketplace across various product ranges paired with almost immediate delivery (Gray, 2018). Together with its further online and offline innovations7 Amazon fundamentally modified customers’ expectations of a purchase experience, which is titled the “Amazon Effect” (Grosman, 2018). Other players like Addidas, Sweetgreen, and Brandless8 9, triggered similar disruptive effects at a velocity traditional retailers have been unable to keep pace with (Hanbury, BI, 26.03.18).
Pressure on Already NarrowMargins
By virtue of the intensifying competition, “margins are declining in almost every retail category” (McKinsey, 2018, p. 49). While scale-effects used to be the most important lever for cost and efficiency, bottom-line improvements today primarily stem from tech-driven efficiency improvements along the value chain (Gray, 2019). Frequently, only emergent, agile firms are found to be equipped with those cutting-edge innovations. According to McKinsey, retailers leading in technology could achieve two to five percentage points higher EBIT’s than less advanced ones, yielding a competitive advantage (McKinsey, 2018, p. 10). Again, to obtain this, significant investments in warehouse automation and advanced analytics are required forboth new and established players (Gray, 2019).
To conclude, technology is not solely revolutionizing the way retailers approach consumers, it is also transforming how they increase productivity along the supply chain. Retailers, and ultimately their owners face the challenge of balancing online and physical store customer experience while sustaining margins despite increased pricing pressures. Hence, it is reasonable to assume that a notable skillset of human capital is required in order to sustain this balance and remain competitive.
2.2 Private Equity Value Creation in Retail
Consequential to record fundraising activities in recent years, the amount of dry powder in the PE industry has accumulated (Füss, Fuchs, Jenkinson, and Morkötter, 2017, p. 6). As the amount of capital chasing targets rose, detecting attractive investment opportunities has evolved into being a major challenge for GPs. Industries like the retail sector thereby increasingly received private equity’s attention during the past two decades (Baker & Kennedy, 2007, p. 92). The following graph lays out PE buyout activity, and investment failure in the retail sector, together with the amount of the overall industries dry powder. Strong capital commitments to the PE sector in the mid-2000s can be observed, leading to an LBO boom in most industries, including retail.10 During the financial crisis, deal activity in retail dropped by 50 percent compared to 2007, while the number of write-offs notably increased. However, merely a few years later, the buyout sector fully recovered and pre-crisis LBO levels were exceeded in 2014. In 2016, notably the year preceding the “retail apocalypse”, the activity peak was reached with 337 retailers being acquired. Thereafter, the number of retail LBOs declined, while writeoffs of PE backed retailers climbed to financial crisis levels. These developments already indicate an amber light for the difficulties faced by the retail private equity sector. Additionally, the graph depicts the inherent economic cyclicality of both PE deal activity and fundraising volumes (see also Kaplan &Strömberg, 2009, p. 126).
Abbildung in dieser Leseprobe nicht enthalten
Exhibit 1: Number of retail LBOs, LBO write-offs in retail, overall dry powder in buyout funds (to 2000-2018) (source: Preqin as of 15.03.2019, own illustration)
The reasons for the observed, significant amounts of capital private equity firms deployed in the retail sector, as well as the characteristics of the individual deals will be assessed subsequently. Following the investment cycle, sourcing an investment is the first action executed by the GP. Retailers large, predictable cash flows stemming from sales volumes, as well as their solid assets bases to use as collateral for debt, turn them into attractive targets for LBOs (Baker & Kennedy, 2007, p. 92). Additionally, the business model of retail companies ordinarily appears tangible and less risky compared to inter alia oil and gas or agriculture businesses, where cash flows heavily depend on commodity pricing (Harding, Tillen & Toner, 2011). As a corollary thereof, GPs have increasingly dedicated funds to retailers. After acquiring the portfolio company, they plan on exiting the investment after approximately five years (Kaplan & Strömberg, 2009, p. 130). Considering the rather short holding periods in private equity11, GPs face a strong imperative on generating short-term returns by theirthree value creations levers: financial engineering, operating improvements and multiple expansion (Achleitner et al., 2011, p. 155). While the latter is often dependent on market conditions, especially financial engineering and operating improvements impact the retail company’s business model and competitiveness.
Financial engineering constitutes a pivotal value creation lever in PE transactions. Most LBOs are financed with 60 to even 90 percent debt, payed back through the company’s free cash flows (Kaplan & Strömberg, 2009, p. 124).12 In the case of Toys”R”Us, its debt burden increased from USD1,86 bn before, to USD5,20 bn after its LBO (DiNapoli & Rucinski, 2017)13. Financial engineering tools, besides leverage, are more specific to the portfolio company’s industry. A key characteristic of established retailers is the considerable value of real estate assets on their balance sheets (Applebaum & Blatt, 2018, p. 58).14 In order to generate additional cash flows, it became common practice for PE firms to dispose of the majority of these real estate assets and yet lease them again in sale-and-lease-back agreements. Hence, retailers owned by GPs usually face the constraints of both interest as well as lease payments (Applebaum & Blatt, 2018, p. 58). This practice has been heavily criticized in academia and practice, since the implemented changes might undermine the retailer’s financial stability, considering the aforementioned pressure on retail margins (Burch & Lawrence, 2012, p. 248). Additionally, due to the fact that most of the cash flow is absorbed by interest and rent payments, the retailer is even less equipped with excess capital to invest in its business and technological improvement. As previously stated, precisely, these investments are vital in order to stay competitive in the sectors current environment (Applebaum & Blatt, 2018, p. 58).
In terms of operating improvements, empirical evidence is somewhat consistent in observing that, on average, GPs increase performance and productivity metrics like sales, operating margins and, Capex management in their portfolio companies (Kaplan, 1989, p. 250; Kaplan & Strömberg, 2009, p. 133). Nevertheless, these results should be treated with caution since the effects of LBOs might vary significantly depending on the portfolio company industry, GP characteristics and skills, as well as macroeconomic conditions (Kaplan & Strömberg, 2009, p. 133). In the retail sector, typical operating improvements include disposing of underperforming product-lines and stores, changes in the supplier structure and other cost cuts in orderto improve bottom-line margins (Burch & Lawrence, 2012, p. 253). Those initiatives require little capital investments, compared to the capital intense technological improvements many successful retailers currently implement to enhance their bottom-line. Furthermore, GPs aim to achieve top-line growth by expanding product categories or regional presences, either organically or through buy-and-build strategies. However, not all of these measures are considered sustainable in the long term. Burch and Lawrence (2012, p. 248), who examine the impact of PE funds involvement in the retail food sector indicate that GPs are “.. .not taking control of supermarkets in order to better manage these in the medium to long term, or to compete more effectively with other retail chains.” GPs ultimately are short-term return driven, pressured to exit their investment after only a few years in order to redistribute capital to their LPs and the investment team (Thaker, 2014, p. 17).15 Thereby, the concern of the private equity implemented operating improvements like cost cuts solely subsisting inthe shortterm is raised (Kaplan & Strömberg, 2009, p. 133).
The GPs investing in retail might underestimate the indispensability of investments in software, big data analytics, and supply chain optimization, since those could be anticipated to only pay off in the medium to long term. As seen in recent years, the disruptive change in the retail sector emerged faster than expected by many players, leaving even established retailers who did not react quickly enough, behind. Simultaneously, PE firms conduct excessive financial engineering despite the already narrow retail margins and thereby, according to Applebaum and Blatt (2018, p. 56), induce the high proportion of private equity-backed retail bankruptcies. This lack of sensitivity to the retail sector’s fast-paced disruption most likely results from a lack of skills and industry knowledge in the GPs investment team (Applebaum & Blatt, 2018, p. 56). Even though the business model of retailers appears to be tangible at first, the disruptive dynamics could lead to the investments of less-skilled GPs performing poorly, or ultimately facing bankruptcy.
3. Literature Review and Development of Hypotheses
If private equity firms operated in perfectly competitive, frictionless capital markets, returns should solely be determined by systematic risk (Diller & Kaserer, 2007, p. 3).16 Hence, no fund specific characteristics like GP experience, size, and specialization should impact its return, as long as they are unrelated to systematic risk. However, due to market illiquidity, the rigidity of fund flows between LPs and GPs17 and the limited number of target companies PE funds face, a high level of market imperfection can be observed (see Diller & Kaserer, 2007, p. 3; Lerner & Schoar, 2004, p. 7). A corollary thereof is that a PE firms’ skills should influence returns on its investments more significantly than, for instance, the skills of public equity fund managers who are investing in liquid, marketable assets drive their respective returns (Thaker, 2014, p. 45). The observed performance differences between the private equity firms thus arise from their heterogenous skillsets. Therefore, it is to examine which GP characteristics and skills significantly influence investment outcomes.
This chapter provides a literature review on selected performance drivers namely, GP experience, the fund size, and the funds’ industry specialization. Thereby, hypotheses on their impact on investment performance as well as investment failure in the retail sector are derived.
3.1 General Partner Experience
Research on the impact of a PE firm’s experience on the investment outcome has grown within the emerging strand of literature on PE performance. Kaplan and Schoar (2005, p. 1803) provide early evidence of a significant positive relationship between PE experience - measured by the sequenced number of funds raised by a firm - and superior returns. Phalippou & Zollo (2005, p. 14), Gottschalg and Phalippou, (2008, p. 25), Aigner et al. (2008, p. 72) as well as Achleitner et al. (2011, p. 155) and Thaker (2014, p. 46) confirm those findings on either fund or deal levels.18 Thus, existing literature is reasonably conclusive about the positive impact of a PE firms experience on overall performance19, yet imprecise about its effects on sector-specific investment outcomes. In orderto derive ahypothesis with respect to retail investments, it is the aim of this section to comprehensively illustrate factors differentiating experienced firms from their inexperienced competitors. The structure follows the three key levers of PE value creation, namely multiple expansion (split in investment entry and exit multiples), financial leverage, and operating improvements (Achleitner et al., 2011, p. 147).
The first lever of value creation in an LBO is initiated by the identification and acquisition of attractive investment opportunities. Due to the aforementioned accumulation of dry powder in the industry, large amounts of capital chasing a few potential targets (Fuss et al., 2017, p. 6). In this increasingly competitive environment, experienced GPs benefit from so-called “superior selection abilities” (Diller & Kaserer, 2007, p. 3) resulting from far-reaching networks and advanced negotiation skills.
Firstly, personal and professional networks of the GPs investment team can be leveraged in the competitive deal sourcing process. Teten and Farmer (2010, p. 34) report that relationships contribute to almost 50 percent of a PE funds investment opportunities located. Thus, it is to assume that experienced firms, exhibiting longstanding networks, are able to evaluate a larger amount of favorable investments relative to their inexperienced peers. In addition to this, Füss et al. (2017, pp. 19,28) provide significant evidence of a GPs experience and professional network, also increasing the probability of the actual transaction closing. Secondly, not solely professional networks but also a firm’s knowledge improve its acquisition abilities. Achleitner et al. (2010, p. 155) reveal a negative relationship between a GPs experience and its investments’ entry EBITDA multiples20, which they attribute to the experienced GPs superior entry price negotiation skills.21 Negotiating low entry prices might convey a key competitive advantage for achieving higher returns in retail buyouts, as bottom and top-line growth are increasingly difficultto archive.
Regarding transaction financing, the effects of leverage still provide a critical value driver, despite their negative effects on the risk of the transactions in inter alia the retail sector (Burch & Lawrence, 2012, p. 248). Mature GPs with longstanding relationships to lenders as well as a proven track-record are regarded to be more trustworthy than their inexperienced peers. Consequently, experienced firms obtain favorable debt conditions together with more flexible loan covenants, which can enhance returns (Thaker, 2014, p. 46). However, benefitting from cheaper debt frequently results in the willingness to finance a transaction with higher leverage levels and ultimately to pay higher purchase prices. This phenomenon is measured by Achleitner et al. (2011, p. 146), who, on the one hand, prove that experienced GPs lever up their investments more, which, on the other hand, is positively related to entry buyout pricing.22 Even though this effect could be offset by the GP’s superior entry/exit negotiation skills and operating improvements, it should not be underestimated (Achleitner et al., 2011, p. 155). Especially in the retail sector where a company’s high leverage can result in losing its competitiveness, high debt levels carry substantial risks.
During recent years, operational engineering achieved by margin improvements and top-line revenue growth has evolved into an indispensable value driver, given the competitive PE market environment in which financial engineering is no longer sufficient to differentiate oneself among other funds (Achleitner et al., 2015, p. 105). Nevertheless, literature discussing the effects of fund characteristics on operating improvements as a sub-category of performance is sparse. One could argue that analogous to the investment entry phase, accumulated skills and networks of experienced GPs boost operational improvements. Especially when it comes to replacing the targets management team, strong networks can be leveraged to find suitable candidates.23 Whilst providing advice regarding the target’s strategy, daily operations, or financing plans, a GP’s networks and knowledge have been proven to be advantageous as well (Kaplan & Schoar, 2005, p. 1793). This might be of particular importance in industries like retail, where investors currently struggle to achieve high returns (Wigglesworth, 2017; Harding et al., 2011). Additionally, experienced GPs can leverage their “superior selection abilities” throughout the holding period of an investment, if add-on acquisitions are realized. This is of high relevance when the portfolio company is following a “buy-and-build” strategy (Diller & Kaserer, 2007, p. 3).24 The buy-and-build strategies are often utilized for investments in the retail sector since organic growth is difficult to achieve due to the market’s maturity. Moreover, the sector faces high potential for further consolidations, resulting from its still prevailing fragmentation (McKinsey, 2018, p. 35).
Finally, experience benefits PE firms in the exit phase of an investment, mainly due to achieving high exit multiples. According to Achleitner et al. (2011, p. 159), less experienced and reputed GPs tend to be willing to accept lower exit prices, especially when approaching the current funds’ maturity. This stems from the GP being confronted with the pressure of quickly proving an exit track record, in order to expeditiously raise a follow-on-fund. Additionally, bidders acquiring a portfolio company from an unexperienced PE firm might claim a discounted price resulting from information asymmetries regarding the company’s actual condition (Achleitner et al., 2011, p. 159).25
To sum it up, those considerations regarding the effects of GP experience across the value creation levers lead to the first hypothesis derived:
H1 a: Experienced GPs exhibit more successful investmentperformances in the retail sector than inexperienced ones.
This thesis additionally examines buyouts exits by its write-off. The particular relevance of this analysis fathoms on the bankruptcies of numerous retail giants. To derive further hypothesis, the relationship between write-offs and buyout and fund performance is examined hereafter. In literature, five main types of PE portfolio company exits have been specified, namely being trade sales, initial public offerings (IPOs), secondary buyouts, buy-backs, and investment write-offs (Cumming & Macintosh, 2003, p. 513). Contrasting to the first four exit types, the write-off typically involves the investment’s failure as the fund decides not to undertake further efforts regarding the portfolio company. Frequently, write-offs become inevitable due to the bankruptcy of the target. Otherwhiles, despite the company is not defunct, its value in the GPs books is written down to zero, which is terminated as a ‘living dead’ investment since the fund does not anticipate further significant cash flows generated by the asset. At a certain point, this write-down is reported by the GP as a failed investment, respectively a write-off (Cumming & Macintosh, 2003, pp. 513-515).
The few studies analyzing write-offs and its determining factors in private equity and venture capital all utilize write-offs as a proxy for poor buyout performance, the same way a negative IRR or low MoMs would be used (see Gompers et al. (2009, p. 833), Hochberg et al. (2007, p. 262) and Schmidt et al. (2010, p. 24)). The reasons that exit channels are used as a performance proxy mainly stem from the limited availability of quantitative performance data-points like IRRs (Hochberg et al. 2007, p. 253). Building on this approach, this thesis utilizes findings on the relationship between fund level characteristics and MoMs/IRRs, to examine the impact of the same fund characteristics on investment write-offs. This approach presumes a negative relationship amongst MoMs, respectively IRRs and the failure of an investment. Phalippou and Gottschalg (2008, p.1760) support this rationale by verifying the positive relationship between the success of an investment, measured by it being exited via IPO or its sale, and the funds’ real performance.26 Hence, assuming a negative relationship between the inverse of the investment’s success measure used by Phalippou and Gottschalg and the funds’ performance in terms of IRR is straight forward and will be used in the further discussion of this thesis. As a GPs experience is argued to be positively related to investment performance, it should yield a negative relationship to the probability of the write-off of an investment, which is why experienced funds should face lower write-off probabilities than less experienced peers. This theory is also vindicated by Gompers etal. (2009, p. 832) forthe venture capital industry, as they define a VC investment’s success by its exit through either an acquisition, a merger, the registration for an IPO or an IPO itself, and use the writeoff as a measure of poor performance. They find experienced VC firms to show a significantly greater percentage of successful exits (Gompers etal., 2009, p. 828).
To examine a potential drawback of this approach, one could argue that investment write-offs constitute not a proxy for performance, but rather a proxy for risk. Due to the positive risk-return relationship, this implies that better performing funds carry higher risk and thereby experience more frequent write-offs than underperforming funds (Aigner et al., 2008, p. 72). In fact, Aigner et al. (2008, p. 72) and Schmidt et al. (2010, p. 24) find experienced GPs to encounter a larger percentage of failed investments than their inexperienced peers.27 Cummin et al. (2006, p. 237) find the same to hold trough for venture funds. These results would turn the subsequent hypothesis Hlb around.
However, when it comes to measuring risk in the PE industry, research has faced considerable challenges due to the illiquidity of the asset class. A corollary thereof is that no significant relationship between a fund’s systematic or idiosyncratic risk and its performance has been observed in the past (see Ljungqvist & Richardson, 2003, p. 24; Phalippou & Zollo, 2005, p. 10.) Therefore, it is questionable to directly extrapolate from a funds risk to a funds’ investment performance or failure. Instead I will make use of the circumstance that the write-off can be used as a proxy for poor performance.
H1 b: Experienced GPs investing in the retail sector exhibit a lower -write-off probability than inexperienced ones.
3.2 Private Equity Fund Size
The impact of a private equity funds’ size on its investments performance in the retail sector will be addressed in this section. Mega funds like the ones of KKR, Bain Capital, and Apollo Global have been heavily criticized for the bankruptcy of their investments in the retail industry (Ault & Unglesbee, 2018). These firms possess the financial strength to acquire particularly large retailers whose bankruptcies are severe forthe economy due to the significant number ofjob losses (Applebaum & Blatt, 2018, p. 56). Therefore, it is crucial to develop a profound understanding of the effects of fund size on investment outcomes in the retail sector. Arguments for both an increase and a decrease in performance and writeoffs inferred by the fund’s size have been specified by literature and moreover, empirical evidence has been partly contradictory. Therefore, this chapter firstly scrutinizes in greater depth which factors lead to funds increasing in size and subsequently examines potential positive and negative consequences of fund size on investment performance.
Foremost, a GPs fundraising success is driven by both macroeconomic and firm-specific variables. Due to the cyclicality of the PE industry, new funds raised tend to be larger in economic boom times28 compared to the ones raised in economic downturns (Kaplan & Schoar, 2005, p. 1817; Gompers & Lerner, 1998, p. 167).29 On the GP specific level, both, the firms experience as well as its past performance in terms of returns have been found to be positively related to its likelihood of raising a new fund and to the size of this fund (Kaplan & Schoar, 2005, p. 1791; Gompers and Lerner, 1998, p. 150). This can be explained by LPs who prefer to commit their capital only to top-performing funds with a longstanding reputation and proven track record (Thaker, 2014, pp. 45-46). Thereby, a “virtuous cycle” is created, as superior skills accumulate in certain tier-one GPs and thereby attract vast amounts of capital from LPs, creating even larger follow-on-funds.
Given the fact that larger fund sizes are resulting from a GP’s superior past performance, it seems reasonable to assume that there is a certain persistence in performance and sizable funds ceteris paribus show above-average investment returns.30 This assumption is supported by the empirical findings of Gottschalg and Phalippou (2008, p. 25), Thaker (2014, p. 61), Ljungqvist and Richardson (2003, p. 26) and Phalippou and Zollo (2005, p. 14) who all observe a significantly positive relationship between a buyout funds’ size and its performance measured by fund or investment level returns. The underlying reasoning of these results is predominantly based on the fact that fund size seems to consolidate certain performance-related GP characteristics like reputation, economies of scale, and skills, which are examined hereafter (Phalippou & Zollo, 2005, p. 14).
Firstly, Füss et al. (2017, p. 19) find large funds to build larger organizations and to employ more investment professionals with diversified backgrounds. This leads to the GP profiting from access to widespread networks, which, as aforementioned, is advantageous in strengthening all of the three value creation drivers. Additionally, it can be argued that large funds, in order to deploy their capital, execute more deals relative to smaller peers and thereby profit from increased learning effects (Phalippou & Zollo, 2005, p. 14). These considerations regarding the positive relationship between size and investment outcome insist parallels to previous arguments supporting the positive impact of a fund’s experience on performance outcomes. Hence, it is inherent that fund size is often used as another proxy for a GPs skill next to the GPs experience, and therefore is frequently hypothesized to be positively related to investment outcome (Thaker, 2014, p. 45; Phalippou & Zollo, 2005, p. 14).
As straight forward as this approach may appear, it should be treated with circumspection since a different strand of literature presents and substantiates contrary results. Although Kaplan and Schoar (2005, p. 1821) and Aigner et al. (2008, p. 71) find a fund’s performance initially increasing with its size, they observe this relationship to be concave, causing the performance of larger funds to decline. They further assess this matter by deriving the “ideal” fund size, which maximizes the fund’s returns (Kaplan & Schoar, 2005, p. 1805)31. This maximum is found to be fairly small, with USD90 MM by Kaplan and Schoar and EUR 24 MM by Aigner et al. (2008, p. 75). Thus, large funds are confronted with “strong diseconomies from scale and scope,” which indicate performance to decline with size (Kaplan & Schoar, 2005, p. 1821). The underlying reasons forthis circumstance are outlined hereafter.
Firstly, GPs frequently struggle in detecting suitable, skilled human capital to source, acquire, and oversee their portfolio companies, especially in senior ranks (Kaplan & Schoar, 2005, p. 1821). Secondly, large funds face the danger of simultaneously investing in an exceptionally high number of portfolio companies. Lopez-de-Silanes et al. (2015, p. 403) prove that investments tend to underperform when they are hold concurrently with too many other portfolio companies due to the limited attention each company obtains.32 It might be the case that especially targets with business models that appearto be tangible and generate steady cash flows like it is the case for retailers, retrieve especially little monitoring, which would lead to poor performance and bankruptcies. Thirdly, only a limited number of attractive investment opportunities exist in a market, which denotes that some GPs could deliberately keep their fund size smaller in order to avoid not bringing their entire capital raised to work (Kaplan & Schoar, 2005, p. 1821; Phalippou & Zollo, 2005, p. 14). Finally, according to Aigner etal. (2008, p. 76) this limited number of attractive investment opportunities in a market can also result in larger funds paying higher entry prices for their investments, as they have no further opportunities to deploy their capital. Following this line of argumentation, a GPs’ skill appears not to be easily scalable by solely investing its increased levels of capital in a greater quantity of portfolio companies or in targets with higher enterprise values.
In light of the pro and contra aspects of fund size, the observed mixed empirical evidence is unsurprising.33 Even though the rationale behind potential negative effects of fund size due to diseconomies of scale appears to be reasonable, evidence on a positive relationship between fund size and performance slightly predominates. I will henceforth follow the approach used by inter alia Thaker, reasoning that “‘skill is the most important factor” (2014, p. 2) in determining success in the private equity industry and skill being proxied by PE experience, but also by PE funds size. Hence, I weight the learning and network arguments in favor of fund size stronger than potential difficulties in scaling up skills as a GP. This leads to hypothesis 2 a:
H2 a: GPs managing largefunds exhibit more successful investmentperformances in the retail sector than GPs managing smallfunds.
Given that the write-off of an investment has previously been identified as a measure of poor performance, it is apparent to assume a negative relationship between a fund’s size and its investment failures in the retail sector. Large, reputable GPs with superior investment selection and value creation abilities can be inferred to face fewer write-offs. This is supported by Achleitner et al. (2011, p. 159) who find a negative relationship between fund size and the leverage used in the acquisition of its portfolio companies while controlling for the GPs experience and the investment size. Since high leverage burdens are identified as a fundamental cause of retail and non-retail bankruptcies, this observation can diminish the risk of the portfolio company’s failure.
H2 b: GPs managing largefunds exhibit a lowerprobability of a retail investment’s -write-off than GPs managing small funds.
3.3 Retail Industry Specialization
With competition increasing in the PE sector, the pressure of gaining a competitive advantage has intensified for buyout firms. A corollary thereof is the approach of narrowing down the strategic focus of the entire firm or certain funds in terms of financing stage, investment geography, or industries (Cressy, Munari & Malipiero, 2007, p. 7).
Regarding the financing stage, a primary distinction is made between venture capital and buyout investments.34 Both approaches vary significantly in their investment rationale, size, risk, and financing structure (Fenn et al., 1995, p. 18). Whereas venture investors deploy equity to young companies in industries with an expected high-growth profile, leveraged buyouts target established players in mature sectors, providing stable cash flows and a significant asset base (Fenn et al., 1995, p. 19). Hence, due to the sole nature of the retail industry, venture investments in the sector are found to be extremely rare (Caldbeck, 2015), and only certain disruptors have successfully raised venture capital in recent years (Pitchbook, 2018). It is to assume that those few venture investments in the retail sector are subject to unique dynamics and success factors, which is why they would require a distinctive analysis. Henceforth, this paper limits its attention to leveraged buyout investments, which is why the specialization regarding financing stages is not of relevance for further discussion.
Additionally, geographic specialization is disregarded since national boundaries for fund flows varnish within this increasingly globalized world. Additionally, literature is reasonably conclusive on the nonexistent relationship between geographic specialization and PE performance (see Aigner etal., 2008, p. 81; Lossen, 2006, p. 35). Given those considerations, a fund’s specialization in terms of portfolio company industries and its impact on investment outcome will be addressed in this section. As in the derivation of the previous hypotheses, arguments supporting a positive relationship between the variables draw on the accumulation of skills, knowledge, and professional networks.
The first main argument in favor of industry specialization positively influencing investment performance is based on the development of sector-specific skills. Originating from literature on organizational learning, learning curve effects arise from steady interactions within a certain context and matter of subject (Argote, 2005, p. 41). Hence, specialized GPs tend to accumulate knowledge in their targeted sectors and PE firms focusing on retail investments should profit from a deeper understanding of the industries complexities, critical success factors and disruptive dynamics (Gompers et al., 2009, p. 820; Cressy et al., 2007, p. 648). Thus, information asymmetries across the entire value creation cycle are mitigated (Lossen, 2006, p. 14). Beginning with deal sourcing, specialized GPs should determine strengths and weaknesses, and thereby the intrinsic value of the target more accurately than their unspecialized peers (Cressy et al., 2007, p. 648). During the investment management phase, specialized GPs are capable of providing more specific advice and effective monitoring to the portfolio company (Cressy et al., 2007, p. 649). For instance, sale-and-lease-back agreements of the retailer’s real estate assets can be highly sophisticated and require expertise, which unspecified funds might not be able to develop (Baker & Kennedy, 2007, p. 94). The same holds true for expertise in e-commerce and omni-channel strategies, as well as supply-chain management (Ault & Unglesbee, 2018). This argument is supported by Cressy et al. (2007, p. 665), who find that industry specialization on the PE firm level increases the mean post-buyout operating profitability of the portfolio company by 8.5 percent.35 This, in turn, is positively related to higher equity returns determined by IRR (Achleitner c7«/..20ll.p. 153).36
A second argument based on the accumulation of industry-specific human capital within a PE firm focusses on beneficial networks, created by repeatedly acquiring portfolio companies in similar industries. As discussed in the previous hypothesis, personal and professional networks have been proven to be advantageous for LBOs in relation to all three value creation levers (Füss et al., 2017, p. 28). Hence, specific networks in the retail industry can be utilized to gain access to more attractive targets and to facilitate a more efficient investment management process (Rigamonti et al., 2016, p. 1421).
1 The firms performing LBOs hereafter are referred to as private equity firms or GPs (Kaplan & Strömberg, 2008). The terms LBO, buyout and private equity deal will be used interchangeably.
2 The most noticeable researchers inter alia are A. Achleitner (TUM), S. Kaplan (Chicago Booth School of Business), J. Lemer (HBS), A. Schoar (MIT), T. Jenkinson and L. Phalippou (Oxford Said Business School), and P. Strömberg (Stockholm School of Economics)
3 There has been a longstanding dissent about whether PE funds actually outperform public markets or not. Kaplan and Schoar (2005, p. 1791) and Phalippou and Gottschalg (2009, p. 1749) find returns gross of fees outperforming stock markets slightly, while returns net of fees underperform. However, the majority of the more recent research like the one ofHarris, Jenkinson and Kaplan (2014, p. 24) finds PE funds outperforming markets and criticizing previously used datasets due to incompleteness and biases.
4 Contrary, investment write-offs and its determinants have been studied more intensively in the venture capital sector, as failure is more prevalent when investing in early stage companies (see Cumming and Macintosh (2003, p. 512), Gompers, Kovner and Lemer (2009, p. 833), Hochberg, Ljungqvist and Lu (2007, p. 270)).
5 Hence, “bricks and mortar” strategies turn into “bricks and clicks” or “omni-channel” shopping, which aims to combine e-commerce withphysical store presence (Gray, 2019)
6 The decreasing brand loyalty of especially younger customers is an additional determinant that is shifting the retail industry (McKinsey, 2018, p. 5).
7 Namely the acquisition of grocer WholeFoods, the establishment of cashier-less AmazonGo grocery stores, as well as its recent addition: physical “Amazon 4-star” stores selling the best rated items (Gray, 2018).
8 For details on the individual, disruptive business models please see Hanbury’s Business Insider article (2018)
9 The gravity of this shift can be seen in US retail giant Walmart planning to hire 2,000 machine learning experts and software engineers in2019 alone (Gray, 2019)
10 This increased activity is referred to as the second buyout wave, after the first one occurring in the late 1980ies (Kaplan & Strömberg, 2009, pp. 127, 138)
11 In contrast to strategic acquirers, who, in most cases, plan to hold the company permanently.
12 The heavy cash flows in retail are not solely attractive for debt repayments, but are also used to reward the GP by dividends and thereby increase returns (Burch & Lawrence, 2012, p. 251)
13 Since the overall buyout value payed accounted USD 6,6 billion, (DiNapoli & Rucinski, 2017), the leverage ratio can be approximated by 78,8 percent (= 5,2bn/ 6,6 bn)
14 Already implied by the term “bricks and mortar” retailer, these assets are constituted mainly by physical stores, office space, and warehouses.
15 According to Thaker (2014, p. 17), GPs face two primary motivations: “to generate high returns, and to generate those returns in as short time as possible.”
16 Phalippou and Zollo (2005, p. 10) evaluate a funds exposure to systematic risk by assessing its beta. They find no significant relationship between systematic risk and a fund’s performance.
17 Ljungqvist and Richardson (2003, pp. 2 ff.) examine the “extreme rigidity” of capital flows between LPs and GPs, finding that, on average, ten years pass until the capital is returned to the LPs in order to generate returns.
18 All of the cited papers follow the Kaplan and Schoar (2005, p. 1803) approach of including the sequenced number of previously raised funds as a proxy for PE firm experience. Frequently, measures like firm age incorporated as further proxies for experience as well.
19 Similar results have been found to hold for the venture capital industry, see Gompers et al. (2009, p. 833) and Proksch, Pinkwart, Schefczyk and Stranz (2016, p. 19).
20 The EV/EBITDA multiple is one of the most frequently used valuation metric, since it accounts for differences in companies capital and corporate structures (Phalippou, 2017, p. 27)
21 Since entry multiples are also strongly affected by a variety of market-related, external factors, Achleitner et al. (2011, pp. 155, 156) control for deal size, as well as for variables related to the industry and macroeconomic environment. Results remained significant.
22 The positive relationship between leverage levels in transactions and entry pricing is explained by Achleitner et al. (2011, p. 157) as it follows: If GPs are obtaining cheaper debt conditions, either due to favorable credit markets or due to their experience, they are able to pay higher entry prices and win the auction process, but meanwhile still realize their targeted returns.
23 Despite the PE’s presence, the target’s management is still responsible for leading the day to day operations and thereby is an essential factor for achieving superior investment returns. This is the reason why GPs often replace the existing executives with, in their opinion, better skilled candidates and align managements interests with their ownby using financial incentives (Phalippou, 2017, p. 49).
24 Buy-and-build strategies are defined as acquisition strategies in which a portfolio company is utilized as a platform for additional acquisitions, which then will be exited as a unified entity (Hoffmann, 2008, p. 34).
25 Fenn, Liang and Prowse (1995, p. 35) support this considerationby observing that experienced GPs are able to achieve higher exit prices due to the buyer’s perception that experienced funds “do not bring lemons to market.”
26 These two exits summarize the first four of the aforementioned ones, since a sale constitutes trade sales as well as secondary buyouts and buybacks. Hence, all exits not being write-offs are used to verify their positive relationship to performance in terms of fund IRR and MoM (Phalippou & Gottschalg, 2008, p.1760).
27 While Aigner et al. (2008, p. 72) use the percentage of portfolio companies with a negative IRR as a proxy for investment failure, Schmidt et al. (2010, p. 24) directly utilize the investment write-off.
28 “Boom times” are proxied by superior Nasdaq/S&P 500 performance by Kaplan and Schoar (2005, p. 1817) or real GDP growth in the previous year, respectively, as by Gompers and Lemer (1998, p. 167).
29 Interestingly, funds raised in boom times tend to perform worse since they often exit their investments not until the economy is facing a downturn which implies lower exit multiples (Kaplan and Schoar, 2005, p. 1792).
30 Kaplan & Schoar (2005, p. 1791) prove that there is a certain level of performance persistence of funds raised by an individual private equity firm.
31 See section 4.5.2 for a description of how the ideal fund size is derived by Kaplan and Schoar and Aigner et al.
32 Lopez de-Silanes et al. (2010, pp. 381, 388) quantify the number of simultaneous investments by computing the average of simultaneous investments held by the GP during every month of the analyzed targets’ holding period. The median investment is found to be hold concurrently with seventeen other investments.
33 In addition to the body of literature suggesting a positive relationship between size and returns, and to the results onthe correlation being negative, Harris etal. (2014, p. 1852) find a mostly non-significant relationship between performance and fund size.
34 Venture investments range from seed to early- and late-stage investments, while LBOs are subdivided in the two main categories ofbuyouts and the turnarounds of already struggling firms (Fenn et al., 1995, p. 17).
35 The dependent variable “operating profitability” is determined by the three-year mean post-buyout return on total assets, while the independent variable is measured by a dummy assessing a firm’s percentage of investments in the sector of each deal, compared to competing firms in the entire sample (Cressy et al., 2007, p. 653).
36 Achleitner et al. (2011, p. 153) proof that an increase in a targets operating performance yields higher exits multiples and thereby significantly higher IRRs.