American Economic Review: Papers & Proceedings 2015, 105(5): 442–446

http://dx.doi.org/10.1257/aer.p20151000

ORGANIZATION, MANAGEMENT, AND

ECONOMIC GROWTH

Do Private Equity Owned Firms Have Better

Management Practices?†
By Nicholas Bloom, Raffaella Sadun, and John Van Reenen*

Private equity (PE) ownership has become Reenen (2015). In summary, we find that PE
commonplace within the United States, and is owned firms are typically well managed. They
increasing its presence across Europe and Asia. have significantly better management practices
However, there is still some debate over the than almost all other ownership groups such
impact of PE acquisitions on firms and the chan- as family-run, founder owned, or government
nel through which these affect organizations. owned firms. The only exceptions are dispersed
Davis et al. (2014) suggest that PE acquisitions shareholder firms (e.g., publicly listed firms)
in the United States increase firm-level produc- and family firms run by external (nonfamily)
tivity by expanding productive plants and con- CEOs, which have similar levels of our man-
tracting unproductive plants, suggesting superior agement score to PE owned firms. This cor-
managerial skills around investment and plant relation is robust to controlling for observable
selection. Similarly, Bernstein and Sheen (2014) aspects of the firm such as size and industry.
examine the effect of restaurant chain buyouts in It also holds both in developed and less devel-
Florida and report significant improvements in oped economies, suggesting PE ownership is
store-level operational practices in chain-owned associated with superior management regard-
stores relative to franchised locations, where less of the particular country in which the firm
presumably private equity owners had more is located. PE ownership is linked in particular
limited influence. However, Smith (2014) uses with improved monitoring and operational prac-
Indian data to suggest that PE selects already tices—the collection and use of data associated
productive firms and provides them with finan- with modern management technologies such
cial support to grow, providing no post-takeover as Lean manufacturing. PE owned firms also
improvement in performance or management. show stronger performance related incentive
Lerner, Leamon, and Hardymon (2012) present practices, but this advantage is smaller. Finally,
a longer discussion of the mixed evidence. PE ownership is also associated with greater
In this paper we peek inside the black box delegation of authority to plant managers, espe-
of PE ownership by examining the association cially in demand related activities such as sales
between PE ownership and management prac- and marketing and new product introductions.
tices. We do this by using a management evalu- One note of caution is that, because of the
ation score developed in Bloom and Van Reenen cross-sectional nature of our data, we cannot
(2007) and extended in Bloom, Sadun, and Van distinguish selection from treatment effects.
That is, the superior management of PE owned
* Bloom: Stanford University, 579 Serra Mall, Stanford,
firms could come entirely from purchasing
CA 94305 (e-mail: [email protected]); Sadun: well-managed firms, rather than improving
­
Harvard Business School, Harvard University, Morgan Hall firms’ management over time. While this is pos-
(e-mail: [email protected]); Van Reenen: London School of sible, we think it is unlikely to fully explain our
Economics, Houghton Street, London, WC2A 2AE (e-mail: results because in the United States and United
[email protected]).
†
Go to http://dx.doi.org/10.1257/aer.p20151000 to visit Kingdom the common perception is that PE
the article page for additional materials and author disclo- firms look for badly managed targets to acquire
sure statement(s).
442
VOL. 105 NO. 5 DO PRIVATE EQUITY OWNED FIRMS HAVE BETTER MANAGEMENT PRACTICES? 443

for performance turnaround, implying negative Table 1—Private Equity Ownership Across Our Sample
selection effects.
All PE PE
Country firms firms share (%)
I. Measuring Management
Argentina 566 4 0.71
We use a “double-blind” management survey Australia 470 15 3.19
developed in Bloom and Van Reenen (2007). In Brazil
Canada
1,145
418
11
33
0.96
7.89
summary—with full details in Bloom, Sadun, Chile 544 8 1.47
and Van Reenen (2015)—we collect informa- China 761 1 0.13
tion on 18 dimensions of firms’ management Colombia 170 3 1.76
grouped into three areas: (i) performance mon- Ethiopia 131 0 0.00
itoring (information collection and analysis);
France 751 36 4.79
Germany 685 29 4.23
(ii) effective targets (using stretching short- and Ghana 107 3 2.80
long-run targets); and (iii) performance incen- Greece 585 7 1.20
tives (rewarding high-performing employees, India 921 2 0.22
Italy 628 20 3.18
and retraining or moving underperformers). In Japan 172 0 0.00
a separate part of the survey, we also collect Kenya 184 3 1.63
information on the extent to which plant man- Mexico 524 5 0.95
agers can make autonomous decisions (i.e., Mozambique 109 5 4.59
without their corporate headquarters approval, Myanmar 146 1 0.68
New Zealand 149 8 5.37
see Bloom, Sadun, and Van Reenen 2012 for Nicaragua 97 1 1.03
details). Nigeria 118 0 0.00
One part of the double-blind methodology is Poland 364 10 2.75
that our interviewers are not told anything about Portugal 410 7 1.71
Republic of Ireland 161 5 3.11
the financial performance of the firms they inter- Singapore 373 0 0.00
view. They are simply given the firms’ names Spain 213 8 3.76
and telephone numbers, making them “perfor- Sweden 377 44 11.67
mance blind” as they generally have not heard Tanzania 150 1 0.67
of the medium-sized companies we survey. The Turkey 332 15 4.52
United Kingdom 1,618 108 6.67
second part of the double-blind technique is that United States 1,516 71 4.68
the managers we interview are not informed Vietnam 76 0 0.00
that they are being scored. To achieve this, we Zambia 68 1 1.47
score management using a predefined practice
Total 15,038 465 3.09
grid provided by a leading international con-
sultancy company and open-ended questions. Note: Private Equity ownership across our random sample
Having interviewers “performance blind” and (50 to 5000 employees) of manufacturing firms. The number
managers “scoring blind” helps to minimize any of firms interviewed in each country (the “All firms” num-
potential bias in the survey. The 18 individual ber) was driven by funding and country size.
management dimensions are averaged into one
overall management score after they have each
been normalized to z-scores (a mean of zero and public and private—with 50 to 5,000 employ-
a standard-deviation of one). ees in each country we survey. The sample of
Over multiple survey waves since 2004 we countries—which is chosen both by economic
have collected management practice scores from size (i.e., we targeted large economics like the
over 15,000 interviews in over 10,000 manu- United States and China) as well as regional rep-
facturing plants across 34 countries, which we resentation—is shown in Table 1. This reports
analyze in this paper.1 These firms are randomly the number of firms in each country, alongside
drawn from the population of all firms—both the number and share that are PE owned. One
striking finding is the spread of PE ownership
1 around the world. While it is unsurprising to
There are multiple interviews of firms, mainly because
we built in a panel element, following the same firms over see over 5 percent of plants owned by PE in
time. A full replication file is available at http://www. Northern Europe and North America, we also
stanford.edu/~nbloom/PE.zip . see PE ownership rates above 1 percent in most
 444 AEA PAPERS AND PROCEEDINGS MAY 2015

0.8
Dispersed 3.20
shareholders 3.05
Management score

0.6 Private equity

2.96
3.13

Family owned, 3.05

external CEO 3.01
0.4 2.95
Managers
2.82

Private 2.83
individuals 2.84
0.2
2.72
Government
2.81

Family owned, 2.69

0 family CEO 2.73
Country and industry controls
1 2 3 4 5 Founder 2.55
Raw data
2.73

PE Dispersed shareholders
2.6 2.8 3 3.2
Family CEO All except PE
Average management scores

Figure 1. The Distribution of Management Scores Figure 2. Average Management Scores Across
Across Ownership Types Ownership Types

Note: The kernel distribution of management practice scores Notes: Management scores for 15,038 firms. Raw data and
for 15,038 firms, of which 465 are owned by PE, 4,076 by with country and three-digit SIC industry controls.
dispersed shareholders (publicly listed), and 2,539 by fam-
ily and have a (second or greater) generation family CEO.

line), dispersed shareholders (dotted line), fam-

of Asia, South American, and even African ily owned and managed firms (hatched line),
countries. So while PE ownership may be rare in and all other firms except PE (­dashed line).
those countries, it still occurs on a regular basis. Three results are clear. First, PE firms have gen-
To validate the accuracy of the manage- erally a superior distribution of management
ment scoring we carry out two pieces of anal- practices compared to other firms. Second, PE
ysis (see Bloom et al. 2014 for details). First, firms have a particularly large advantage over
we ­reinterview 222 firms using both a different family firms, a group we focus on because they
interviewer and a different plant manager at the represent the most common target group for PE
same firm, finding scores have a correlation firms in developing countries. Third, PE firms
of 0.51 ( p-value < 0.001). This suggests our have similar management practices to dispersed
management scores are consistently measur- shareholder (publicly listed) firms, a group
ing firm-level practices. Second, we match our which is frequently the source of acquisitions in
management practice data to firm-level perfor- developed countries like the United States and
mance indicators from independently collected United Kingdom.
company accounts, such as productivity, prof- In Figure 2 we show the ranking of PE firms
itability, sales growth, and Tobin’s Q. We find by management practices against a wide range
that better management practices are strongly of ownership types. The solid black bars plot
correlated with these independently collected this data using the raw management scores. In
firm performance measures in every region we the raw data, apart from dispersed shareholders
interviewed. For example, a one unit increase (e.g., publicly listed firms), PE ownership tops
in our management practice score is associated the ranking. The gray bars just below plot this
with a 15 percent increase in productivity, a 34 data after controlling for country and industry
percent increase in size, and a 4 percent increase dummies. A similar ranking persists even within
in employment growth rates. the same country and industry, although PE
owned firms now appear to be slightly below
II. Management Practices in family firms run by external (i.e., nonfamily)
Private Equity Firms CEOs (the difference between the two is not
significant).
In Figure 1 we plot the distribution of man- To further investigate the management prac-
agement practices across PE owned firms (solid tices advantage of PE firms Table 2 runs a
VOL. 105 NO. 5 DO PRIVATE EQUITY OWNED FIRMS HAVE BETTER MANAGEMENT PRACTICES? 445

Table 2—Management Practices Score By Ownership and skills, plus a set of survey noise controls like
Type Compared to Private Equity interviewer dummies and interview duration
controls. While some of the management gaps
Dependent variable
Management score (1) (2) (3) between ownership types shrink somewhat—
mainly because we have controlled for firm size
Founder −0.854*** −0.480*** −0.323*** and employee skills which are partly outcome
(0.046) (0.045) (0.039)
variables (well managed firms are likely to grow
Family CEO −0.652*** −0.430*** −0.320*** faster and be able to hire more skilled employ-
(0.047) (0.045) (0.039)
ees)—we still see PE firms have significantly
Family ownership, −0.127** 0.020 −0.034
external CEO (0.055) (0.052) (0.045)
higher management scores than most other own-
ership groups.
Dispersed 0.097** 0.135*** 0.061
shareholders (0.045) (0.043) (0.037) Finally, Table 3 examines PE ownership by
Private individuals −0.449*** −0.237*** −0.137***
type of management practice, breaking this into
(0.046) (0.044) (0.038) the three groups we outlined above: monitoring,
Government −0.619*** −0.284*** −0.322*** targets, and incentives. For each of the three areas
(0.068) (0.068) (0.060) we compare PE owned firms to all other firms
Other −0.276*** −0.099** −0.102** including the same set of country, industry, firm,
(0.053) (0.050) (0.044) and noise controls used in Table 2. We find that
Managers −0.272*** −0.242*** −0.119* PE firms appear to have a particularly large gap
(0.075) (0.071) (0.063) in monitoring practices. These are the type of
“Lean” manufacturing practices around contin-
Observations 15,038 15,038 15,038
uous performance measurement, improvement,
Country and No Yes Yes and feedback that originates in world-class firms
industry dummies
like Toyota. Table 3 shows that PE firms are also
Firm and noise No No Yes
controls
relatively strong at setting effective targets, which
involves setting stretching but realistic targets
Notes: Dependent variable is the management z-score (mean across the whole firm, with these targets match-
zero and standard deviation of one), with PE ownership the ing up in the short and long run. Finally, while
omitted category. Standard errors clustered at the company unconditionally PE firms do have better incen-
level. Industry controls are three-digit SIC industry dum-
mies. Firm controls are employment, age, and the propor- tives over linking pay, promotion, and continued
tion of employees with a degree. “Noise” controls include employment to the effort and ability, when we
the duration of the interview, an interviewee dummy, and an include a full set of country, firm, industry, and
interview reliability score. noise controls, the difference with other owner-
ship becomes insignificant.
In the last two columns of Table 3 we exam-
regression of management practices on owner- ine whether PE firms also differ in terms of the
ship type and an increasing number of controls. decision making authority allocated to the plant
In column 1 we regress management practices manager. This decentralization index records
on a set of ownership dummies with PE as the the extent to which the plant manager can make
omitted base. We find that PE owned firms have autonomous decisions in terms of hiring and
significantly higher management scores than investment, sales and marketing, and product
every other type of ownership group apart from introduction initiatives. We find that PE firms
dispersed shareholders. In many cases this gap appear to be more decentralized relative to other
is large—for example, PE firms have a raw man- ownership types in terms of decisions related
agement gap with family owned, family CEO to sales/marketing and product introduction
firms of 0.652, which based on the association (column 5), while the difference is insignifi-
between management and performance quoted cant with respect to hiring and capital decisions
above, would be associated with about a 10 per- (column 4).2
cent productivity gap and a 3 percent growth
rate gap. In column 2 we include a set of country
and industry dummies and our ownership rank- 2
In other work (Aghion et al. 2015) we show evidence
ings are pretty stable. Finally, in column 3 we that plant manager autonomy (especially in terms of sales/
include a full set of firm controls for size, age, marketing and product introduction) played a key role in
446 AEA PAPERS AND PROCEEDINGS MAY 2015

Table 3—PE Management Practice Gap by Type of Practice and Decentralization

Dependent Decentralization: Decentralization: sales,

variable Operations Targets Incentives hiring and investment marketing, and new products
(1) (2) (3) (4) (5)

Private equity ownership 0.147*** 0.090** 0.027 −0.020 0.114**

(0.037) (0.037) (0.043) (0.037) (0.058)

Notes: All columns include country, industry, firm, and noise controls. Standard errors clustered by firm. Columns 1 to 3 are
based on 15,038 observations. Columns 4 and 5 are baed on 11,598 observations.
*** Significant at the 1 percent level.
** Significant at the 5 percent level.
  * Significant at the 10 percent level.

REFERENCES Bloom, Nicholas, Raffaella Sadun, and John Van

Reenen. 2015. “Management as a Technol-
Aghion, Philippe, Nicholas Bloom, Raffaella ogy?” http://web.standard.edu/~nbloom/MAT.
Sadun, and John Van Reenen. 2015. “Never pdf.
Waste a Good Crisis? Growth and Decentral- Bloom, Nicholas, and John Van Reenen. 2007.
ization in the Great Recession.” http://web. “Measuring and Explaining Management
stanford.edu/~nbloom/absv.pdf. Practices Across Firms and Countries.” Quar-
Bernstein, Shai, and Albert Sheen. 2014. “The terly Journal of Economics 122 (4): 1351–
Operational Consequences of Private Equity 1408.
Buyouts: Evidence from the Restaurant Indus- Davis, Steven J., John Haltiwanger, Kyle Handley,
try.” Unpublished. Ron Jarmin, Josh Lerner, and Javier Miranda.
Bloom, Nicholas, Renata Lemos, Raffaella Sadun, 2014. “Private Equity, Jobs, and Productivity.”
Daniela Scur, and John Van Reenen. 2014. American Economic Review 104 (12): 3956–
“JEEA-FBBVA Lecture 2013: The New 90.
Empirical Economics of Management.” Jour- Lerner, Josh, Ann Leamon, and Felda Hardymon.
nal of the European Economic Association 12 2012. Venture Capital, Pricate Equity, and the
(4): 835–76. Financing of Entrepreneurship. New York:
Bloom, Nicholas, Raffaella Sadun, and John Van John Wiley and Sons.
Reenen. 2012. “The Organization of Firms Smith, Troy D. 2014. “Private Equity Investment
Across Countries.” Quarterly Journal of Eco- in India: Efficiency vs Expansion.” h­ ttp://web.
nomics 127 (4): 1663–1705. stanford.edu/%7Etdsmith6/JMP%20-%20
TDSmith.pdf.

moderating the negative impact of the Great Recession on

firm growth.