how much does management matter?, nicholas bloom

How much does management matter?

Nick Bloom (Stanford)working with Raffaella Sadun (HBS) & John Van Reenen (LSE)

OECD-NBER, September 26th 2014

Long debate over the importance of management for growth and development

Francis Walker wrote a paper called in 1887 in the Quarterly Journal of Economics called “The Sources of Business Profits”

Walker argued that management was the key driver of differences in firm performance

“No potential driving factor of productivity has seen a higher ratio of speculation to empirical study”

- Chad Syversson (2011, Journal of Economic Literature)

But there is still a wide debate – many people claim management is all “hot air” or “BS”

1) Measuring management

2) Impact of management on performance- Regression results- Field experiments

I will try to summarize 10+ years research in 2 areas

World Management Survey has covered about 20,000 manufacturing firms globally since 2004

1) Developing management questions• Scorecard for 18 monitoring and incentives practices in ≈45 minute phone interview of manufacturing plant managers

2) Getting firms to participate in the interview• Introduced as “Lean-manufacturing” interview, no financials• Official Endorsement: Bundesbank, RBI, PBC, World Bank etc.

3) Obtaining unbiased comparable responses, “Double-blind”• Interviewers do not know the company’s performance• Managers are not informed (in advance) they are scored

Basic survey methodology – 3 key steps

Score (1): Measures tracked do not indicate directly if overall business objectives are being met. Certain processes aren’t tracked at all

(3): Most key performance indicators are tracked formally. Tracking is overseen by senior management

(5): Performance is continuously tracked and communicated, both formally and informally, to all staff using a range of visual management tools

Example monitoring question, scored based on a number of questions starting with “How is performance tracked?”

Examples of performance metrics – Car Plant

Examples of a performance metrics – Hospital

9

Examples of performance metrics – Retail (Ctrip)

Examples of performance metrics – Heathrow T5

Example of no performance metrics: Textile Plant

Score (1) People are promoted primarily upon the basis of tenure, irrespective of performance (ability & effort)

(3) People are promoted primarily upon the basis of performance

(5) We actively identify, develop and promote our top performers

Example incentives question, scored based on questions starting with “How does the promotion system work?”

Examples of performance reviews – Retail Bank

14

16

Wide spread of management in manufacturing

N=84N=100

N=58N=108N=130

N=102N=152

N=180N=840

N=1150N=569N=755N=269N=322N=107N=611N=160

N=307N=136N=150N=323

N=364N=523

N=455N=313N=632N=1208

N=412N=403

N=658N=176

N=1289

2 2.5 3 3.5Average Management Scores, Manufacturing

MozambiqueTanzania

ZambiaGhana

EthiopiaNicaragua

KenyaColombia

IndiaBrazil

ArgentinaChina

GreeceTurkeySpainChile

Republic of IrelandPortugal

Northern IrelandNew Zealand

SingaporePolandMexico

AustraliaItaly

FranceGreat Britain

CanadaSweden

GermanyJapan

United States

Africa

Asia

Australasia

Europe

Latin America

North America

Average manufacturing management scores across countries are strongly correlated with GDP

Ethiopia

Ghana

Kenya

Mozambique

Tanzania

Zambia

Australia

New Zealand

China

India

Japan

Singapore

France

Germany

Greece

Italy

Poland

PortugalRepublic of IrelandSpain

Sweden

TurkeyArgentinaBrazilChile

Colombia

Mexico

Canada

United States

22.

53

3.5

Ave

rage

man

agem

ent p

ract

ices

7 8 9 10 11Log of 10-yr average GDP based on PPP per capita GDP(Current int'l $ - Billions)

AfricaAustralasia

Asia

EuropeLatin America

North America

management x log of GDP PPP per capita

Note: April 2013, World Economic Outlook (IMF) indicator

Management also varies heavily within countries0

.51

1.5

0.5

11.

50

.51

1.5

0.5

11.

5

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

Argentina Australia Brazil Canada Chile

China France Germany Greece India

Italy Japan Mexico New Zealand Poland

Portugal Republic of Ireland Sweden United Kingdom United States

Also been looking at other sectors: hospitals

Randomly surveyed population of hospitals in each country that offer acute care (take emergencies), and have an orthopaedics and/or cardiology department. Total of 1687 hospitals.

1.9

2.2

2.4

2.5

2.5

2.6

2.7

2.8

3.0

1.7 1.9 2.1 2.3 2.5 2.7 2.9 3.1Average Management Score

India

Brazil

France

Italy

Canada

Germany

Sweden

UK

US

Again see a very wide spread in hospitals

Source: www.worldmanagementsurvey.com

0.5

11.

50

.51

1.5

0.5

11.

5

1 2 3 4 5 1 2 3 4 5 1 2 3 4 5

1 US 2 UK 3 Sweden

4 Germany 5 Canada 6 Italy

7 France 8 Brazil 9 India

Graphs by Country

Also been looking at other sectors: high-schools

2.94

2.82

2.80

2.78

2.55

1.99

1.72

1 1.5 2 2.5 3 3.5

Management Score

UK

Sweden

US

Canada

Germany

Italy

India

School Management Scores, by Country

91

77

138

287

n

232

127

323

Randomly surveyed population of high schools in each country with 100+ pupils aged 15.

So we find a huge spread in management practices across firms and countries….

….but does this matter?

1) Measuring management

2) Impact of management on performance- Regression results- Field experiments

I will try to summarize 10+ years research in 2 areas

Management score decile

Pro

duct

ivity

Pro

fit

Out

put g

row

th

Exp

orte

rs

R&

D p

er e

mpl

oyee

Pat

ents

per

em

ploy

ee

These management scores are positively correlatedwith firm performance

These management scores are positively correlatedwith firm performance – even with many controls

This positive correlation with performance has been pretty much true in every sector examined

For example, higher management scores correlated with:

• Hospitals: Higher case-mix survival from heart-attack & surgery

• Schools: Better test scores

• Retail: Great profits and sales

• Universities: Higher research rankings

Of course this correlation may not be causal.

So I’ve also been working with the World Bank on management randomized control trial

1) Measuring management

2) Impact of management on firm and national performance- Regression results- Field experiments

I will try to summarize 10+ years research in 2 areas

Took 28 large textile plants near Mumbai and randomized into treatment (improved management) & control (same as before)

Inventory Control: Before

Inventory Control: After

Factory operations: Before

Factory operations: After

Spare parts: Before

Stores: After

Stores: After

Factory information: Before

Factory information: After

8010

012

014

0

-15 -10 -5 0 5 10 15 20 25 30 35 40 45 50 55 60

These simple management improvements increased productivity by 20% within 1 year alone

Control plants

Treatment plants

Weeks after the start of the management experiment

Pro

duct

ivity

(out

put p

er w

orke

r)

Source: Bloom, Eifert, Mahajan, McKenzie & Roberts, forthcoming Quarterly Journal of Economics

Combining this data we estimate management accounts for about 25% of the differences in productivity across firms & countries

Country Share of TFP gap with the USdue to Management

US baselineSweden 7.8%Japan 10.4%Canada 22.4%Great Britain 36.5%Italy 47.7%France 38.7%Brazil 16.9%China 14.9%Argentina 20.6%Portugal 48.2%Greece 32.4%Unweighted av. 25%

Source: Bloom, Sadun & Van Reenen (2014), “Management as a technology”

So why don’t firms just improve their management practices – one reason is they cannot self assess

We asked at the end of the survey:

“Excluding yourself, how well managed would you say your firm is on a scale of 1 to 10, where 1 is worst practice, 5 is average and 10 is best practice”

In fact we formally investigated if firms can self-assess their management practices….

0.1

.2.3

.4D

ensi

ty

1 2 3 4 5 6 7 8 9 10Self scored management

…and found firms are too optimistic on management

“Average”

…and self-scores show no link to performance-2

00

2040

60P

rofit

s (R

OC

E)

0 2 4 6 8 10Self scored management

So how can we work to improve management – in our research we find four key factors:

1) Education (of both managers and non-managers)

2) Free markets (competition and strong legal systems)

3) Professional management (not family-firm CEOs)

4) Light government regulation (notably labor regulations)

Further reading – good recent summary

More research, policy briefs & media available here www.worldmanagementsurvey.com

MY FAVOURITE QUOTES:

Interviewer: “How many production sites do you have abroad?Manager in Indiana, US: “Well…we have one in Texas…”

Americans on geography

Production Manager: “We’re owned by the Mafia”Interviewer: “I think that’s the “Other” category……..although Iguess I could put you down as an “Italian multinational” ?”

The difficulties of defining ownership in Europe

Interviewer : “Do you export any of your products?”

Factory Manager: “No, our products only cater for tastes in ourlocal market”

German Sex Toy Manufacturer

MY FAVOURITE QUOTES:

Interviewer : “Do staff sometimes end up doing the wrong sortof work for their skills?

NHS Manager: “You mean like doctors doing nurses jobs, andnurses doing porter jobs? Yeah, all the time. Last week, we hadto get the healthier patients to push around the beds for thesicker patients”

Don’t get sick in Britian

MY FAVOURITE QUOTES:

Don’t do Business in Indian hospitals

Interviewer: “Is this hospital for profit or not for profit”

Hospital Manager: “Oh no, this hospital is only for loss making”