Insurance Practice

2019 Global Insurance

Pools trends and
forecasts: Distribution
Brokers and agents remain important, but the direct channel is growing.
An in-depth examination of distribution highlights the variance by region and line
of business.

May 2019
 Our 2019 analysis of McKinsey’s Global Insurance Pools database offers detailed trends and forecasts on
the insurance industry. Overall, the global industry grew by more than 4 percent in 2017, the same level as its
compound annual growth rate from 2010 to 2016, and total premiums reached €4.7 trillion. This report, one
in a three-part series, provides analysis and insights on the distribution function, with an in-depth look by
region and line of business.
The distribution model has continued to evolve as
insurers try to better connect with their customers.
Over the years, the industry has been dominated by
face-to-face selling through agencies and brokers.
More recently, however, insurance has seen robust
growth in direct channels. Notable developments
are the prevalence of direct distribution in Europe,
player-level performance in direct motor, and
growth in aggregator revenues.
Distribution in life
In life insurance, the distribution landscape varies
across regions. The distribution mix across regions
reflects trends in local markets and consumer
preferences, since insurers work with distributors
to tailor products and a sales-force strategy to align
with the local environment. We see a preference
for brokers and independent financial advisers in
the Americas while Europe, the Middle East, and
Africa (EMEA) rely more on bancassurance. In the
Asia–Pacific (APAC) region, the traditional tied
agency channel is still more dominant, as people
prefer personalized advice (Exhibit 1). Countries
selected for each region in Exhibit 1 represent the
largest countries driving the regional trends and
distribution mix.
A more detailed look at selected countries reveals
significant distribution trends.
In the US life insurance market, the direct channel
has been gaining shares since 2012. This channel
is increasing in all customer segments as insurance
becomes more digital. Since digital channels are
more prevalent in life and protection products,
a decrease in individual annuities is matched by a
decrease in the share of brokers and agents.
In Mexico, the broker channel’s share increased
from 5 percent in 2012 to 17 percent in 2016, mainly
2 2019 Global Insurance Pools trends and forecasts: Distribution
because some large brokers entered the market for
both group and individual life products.
In Italy, life insurance distribution is fueled largely
by banks, although the bancassurance model has
evolved in recent years, and some agreements
have been renegotiated. From 2011 to 2015, the
profitability of investment products was low while
insurance products offered attractive returns at
relatively low risk. This growth of the bancassurance
channel in life insurance also partly reflects the rise
of the life insurance industry in Italy.
In Japan, bancassurance declined significantly
from 2012 to 2016. The primary cause was the
2013 market decline, which lowered the guarantee
rate for lump-sum whole life insurance (the main
product for bancassurance). And the Bank of
Japan’s introduction of the negative interest policy
in 2016 sparked a further decline in the interest rate
of government bonds. Hence, long-term savings
products were not an attractive option for asset
liability management.
In addition, low interest rates in Japan led life
players to concentrate on foreign currency–
denominated saving products, which benefit from
relatively higher interest rates in Australia and the
United States. However, the Financial Services
Agency, Japan’s regulatory body for financial
institutions, has strengthened supervision of life
players to protect individual customers; it has also
introduced fiduciary principles that require banks
to disclose sales commissions for bancassurance.
Thus, the sale of insurance products through banks
has continued to decline.
Distribution in P&C
The same regional patterns for life insurance
products also hold for property and casualty (P&C):
 Distribution Report
Exhibit 1 of 5

Exhibit 1

Life distribution channel preference varies by region.

Premiums, %¹

Tied agents and branches Brokers Bancassurance Direct and other

Americas Regional total Canada Mexico United States

29 32 28
33 35 36
55

86
48 49
50 50
58 61 17

11 23 11
8 5 8
10 12 4 10 12
6 3 4 5
2012 2016 2012 2016 2012 2016 2012 2016

EMEA Regional total France Italy Spain

23 21 23 21 27 23 23 20

11 11 6 7
16 15
20
24

58 63 64 72 73
53 62
49

5 5 3 4
1 1
2012 2016 2012 2016 2012 2016 2012 2016

APAC Regional total China Japan Singapore

37
50 49 53 46
54 54 56

21
8 11 1 1 16
10
20
39 36 42 45 35 38
34
23
3 4 2 2 2 1 3 4

2012 2016 2012 2016 2012 2016 2012 2016

¹ Figures may not sum to 100%, because of rounding.

Source: McKinsey Global Insurance Pools

2019 Global Insurance Pools trends and forecasts: Distribution 3

 Distribution Report
Exhibit 2 of 5

Exhibit 2

P&C distribution channel preference varies by region.

Premiums, %¹

Tied agents and branches Brokers Bancassurance Direct and other

Americas Regional total Canada Mexico United States

14 22
33 32 35 32
44

58
64
59 65 33 59
58 58
32
19
9 22 7 9
8 13 3 5 7
2012 2016 2012 2016 2012 2016 2012 2016

EMEA Regional total Germany Italy Spain

55 48 48
56 57 56
85 82

27 27
25 25 27
28
9 13 14
8 7 7 8
7 4
10 11 8 11 3 13 11
5 6
2012 2016 2012 2016 2012 2016 2012 2016

APAC Regional total China India² Japan²

39 36 29
42
57 51
24
21 21 95 95
14 6
6
16 20
3 2 39
44 37 41
23 26
5 5
2012 2016 2012 2016 2012 2016 2012 2016

¹ Figures may not sum to 100%, because of rounding.

2
In China, “direct and other” represents both direct and other, while in India it primarily represents direct distribution channel.
Source: McKinsey Global Insurance Pools

4 2019 Global Insurance Pools trends and forecasts: Distribution

 the Americas lean toward brokers and independent
financial advisers, EMEA prefers the agency
channel, and many APAC countries favor direct
sales (Exhibit 2).
Selected countries demonstrate the significant
distribution trends that have emerged.
In the United States, from 2012 to 2016, the agency
channel declined significantly while brokers
McKinsey Global Insurance Pools 2019
continued to gain share. Agents once served as
Distribution Report
the front line in risk selection and pricing, but
Exhibit 3 of 5
advances in predictive models are making this role
obsolete. And while the agent used to be the face
of the brand, now customers increasingly connect
with their carriers through multiple channels. In the
broker space, major acquisitions from the larger
brokerages have increased that channel’s share.
In Germany, the direct channel has gained market
share primarily because of the growth of motor
insurance. Consumers can purchase these products
easily without the need for advice, making them
a natural fit for the direct channel. In addition,
price competitor websites, which already account
for around 3 percent of in-force motor business,
facilitate online purchases.
In China, P&C distribution is dominated by remote
channels, which include online direct, call centers,

Exhibit 3

Direct channel: Growth has slowed in Western Europe but accelerated significantly in
Eastern Europe.

Direct motor GWP¹ Europe, €, billions

CAGR, % per annum

Eastern Europe²
+22
+5
+49 1.1
0.9
0.7 0.6 0.6 0.7
0.6 0.6
0.3 0.4
0.2

2003 ’05 ’06 ’07 ’08 ’09 ’10 ’11 ’12 ’13 ’14 ’15 ’16 ’17E

Direct motor
market share, % N/A N/A N/A 2 4 5 7 7 7 6 6 6 7 7

Western Europe³
+3
+6
+3
+11
17.4 17.9 18.5
15.6 16.5 16.7
14.8
11.5 11.3 12.0 12.0 12.8
10.4
8.4

2003 ’05 ’06 ’07 ’08 ’09 ’10 ’11 ’12 ’13 ’14 ’15 ’16 ’17E

Direct motor
market share, % 9 11 12 12 13 13 14 15 16 17 17 17 18 18

Hard markets Soft markets Hardening Return of

soft markets?
1
Gross written premiums.
2
Eastern Europe includes Czech Republic, Poland, Russia, Slovakia, and Slovenia.
3
Western Europe includes Belgium, France, Germany, Italy, Netherlands, Portugal, Spain, and United Kingdom.
Source: National statistics, McKinsey Global Insurance Pools

2019 Global Insurance Pools trends and forecasts: Distribution 5

 and branch sales. Traditionally, call centers were the
major contributor. More recently, online direct has
grown rapidly thanks to both innovative pure-digital
insurers and sales through digital partnerships
focused on specific use cases (such as third-party
platform direct sales or redirecting leads).
Two notable emerging trends
In most regions, the direct channel has enjoyed
robust growth resulting from higher internet
penetration and consumer preferences toward
convenience and price. An examination of motor
insurance across regions reveals interesting
variations. Western Europe has recently endured
stagnation in the direct channel, but Eastern Europe
continues to exhibit steady growth in direct motor
sales (Exhibit 3).
McKinsey Global Most
Insurance Pools 2019
Eastern European markets have seen
Distribution Report
the direct channel gain market share. In Russia,
Exhibit 4 of 5 the strong growth of direct motor results from
mandatory motor third-party liability (MTPL).
Starting on January 1, 2017, all insurers selling MTPL
also have to make these products available for
purchase online (so-called eOSAGO).
In motor insurance, most UK and German direct
players have outperformed their markets in both
premium growth and profitability as measured by
the combined ratio (Exhibit 4).
Western Europe’s direct channel’s growth is
expected to slow in the future. Some regions are
already starting to exhibit this pattern, and the
profitability of direct channels may explain why.
Some Spanish and Italian players, for example,
have seen no positive impact from their direct
distribution efforts.
Many European insurers have embraced
aggregators as an effective way to reach consumers.
The online price aggregator channel has progressed
significantly over the past decade. As European
aggregators have evolved, they have helped insurers
dramatically expand their customer reach—at a
predictable cost. In almost every European market,

Exhibit 4

Especially in Germany and the United Kingdom, direct players are increasingly able to
outperform their markets.

Motor direct positioning vs local market, advantage vs local market, 2015–16

COR¹ vs market

30 Each dot represents

Managing for Leading the a direct player in the
20
profitability market following countries:
10
18 27
France
0
Germany
–10 Italy
–50 Spain

–60 United Kingdom

% of players in
–70
each quadrant
–80
Investing in
–90 What happens growth at what
18 next? price? 37
–100
–30 –25 –20 –15 –10 –5 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75

GPW² growth vs market

1
Combined operating ratio.
² Gross written premiums.
Source: National statistics, McKinsey Global Insurance Pools

6 2019 Global Insurance Pools trends and forecasts: Distribution

 Distribution Report
Exhibit 5 of 5

Exhibit 5

Aggregators—growth in revenues is slowing down.

Aggregator market share in direct motor, €, millions

CAGR, % per annum

France (30) Germany (36) Italy (48)

400 +3 2,000 1,000

+10

300 1,500 +4 750

+16

200 1,000 +19 500 +70

100 500 250

0 0 0
2007 2017E 2007 2017E 2007 2017E

Netherlands (10) Spain (22) United Kingdom (52)¹

+10 +4
200 600 6,000
+8
150 450 4,500
+25
+10 +40
100 300 3,000

50 150 1,500

0 0 0
2007 2017E 2007 2017E 2007 2017E

1
UK numbers are in £ due to foreign exchange volatility across years.
Source: McKinsey Global Insurance Pools

aggregators enjoyed double-digit CAGR for motor
gross written premiums (GWP) from 2007 to 2017
(Exhibit 5).
Even though the industry is currently dominated by
agents and brokers, both life and P&C insurance
segments have seen direct channels make progress
overall. Current trends suggest a pronounced shift
Copyright © 2019 McKinsey & Company. All rights reserved.
Photography: ©PASIEKA/Getty Images
toward an approach that includes both traditional
and direct channels in the coming years.
Our distribution analysis indicates that growth
in the direct channel has been slowing down in
recent years. Western Europe especially has
seen a significant slowdown in direct, though
growth in Eastern Europe continues to rise. The
global insurance industry still offers tremendous
opportunities for companies that can quickly
identify pockets of value and respond accordingly.

2019 Global Insurance Pools trends and forecasts: Distribution 7

 This report analyzes trends in the distribution function of the insurance industry. We’ve used
historical data to examine current trends and present an assessment based on our understanding
and research. This report is meant to provide a perspective on the direction the industry is
headed. It identifies current trends and key pockets of growth; it does not present a bias or draw
conclusions about any industry, product, or geography. Any forecasts in this report are based on
our own research; they are merely meant to indicate the direction of the industry.

An overview of Commercial Lines Pools Database consists

of information on the P&C commercial insurance
McKinsey’s Global industry with cuts for 17 industries, seven customer
Insurance Pools segments, and six business lines in 66 markets.

McKinsey’s Global Insurance Pools (GIP)
consists of six proprietary databases: Markets
Database, Local Insurers Database, Global
Insurers Database, Commercial Lines Pools,
Direct Distribution Database, and Multi-Access
Database. The data and insights have been
provided by local researchers and regional and
functional experts.
Markets Database contains more than 150,000
data points covering 66 countries, 14 subregions,
and 99 percent of global insurance premiums.
It includes key financial indicators for every
market from 2000 to 2017¹ and projections to
2025. Future numbers prepared using a uniform
forecasting methodology have been validated
by local experts.
Local Insurers Database provides domestic
premium-level data for total, life, and nonlife
businesses of the top insurers in 43 countries.
Additionally, the database also provides more
granular (local) financials, such as cost and
profitability metrics, for 14 countries.
Direct Distribution Database features key
financial information and trends in the life and
nonlife businesses of the major direct channel
players in 11 markets from 2000 to 2017.
Multi-Access Database offers comprehensive
insights into customer multichannel journeys via
analyses of clients’ channel preferences and usage
regarding core insurance products for 13 countries.
How GIP can help support clients
McKinsey’s Global Insurance Pools can help
insurers along several dimensions. GIP’s Granularity
of Growth analysis can identify a company’s specific
drivers of growth; the tool can also benchmark the
company’s growth and profitability against market
performance and competitors and identify the
impact of different macroeconomic scenarios on
growth and future market shares. McKinsey offers
a subscription to the database that gives users
unlimited access to all data.

Global Insurers Database captures group-

level company financials for more than 100
global insurers, including premiums, cost, and
profitability metrics from 2005 to 2017.

1
Currently, the Markets Database has data available for 2017 only for markets in which regulators have already released this information.

8 2019 Global Insurance Pools trends and forecasts: Distribution

Appendix Insurance distribution through branches implies
that the salesmen are part of the insurer’s staff; in
other words, they receive a salary, not commission.
McKinsey’s GIP initiative uses a bottom-up
approach to size insurance markets. The level of
Brokers are independent insurance distributors
detail in our GIP database varies from market to
who are not tied to any company or salaried. They
market. For less advanced markets, the data might
represent a client (not a company) and distribute
include gross premiums written, technical reserves,
products from a panel of companies.
and profits. For more advanced markets, GIP
includes complete sets of financial indicators for
Bancassurance involves distribution through
each product line, including the mix of distribution
bank branches.
channels. Historical data covers the period from
2000 to 2017 and forecasts to 2025.
Direct channels refer to insurance distribution
through remote channels such as telephone,
GIP distinguishes five product groups in life, based
internet, or mail. Other channels include channels
on European terminology: term life, endowments,
not included in any of the above categories (such as
annuities, unit-linked, and group life (see below for
retailers, car dealers, worksite marketing, affinities).
detailed descriptions).

Life product definitions

P&C consists of five product groups: motor, fire
Term life: all types of protection products with purely
and property, liability, accident, and other (such
biometric risk coverage.
as travel). Health is considered a separate line and
consists of all health premiums underwritten by
Endowments: all individual life-savings products
pure health insurers and life or P&C insurers (based
(both single and regular premium) that provide a
on data extracted either from the life data or P&C
guaranteed credited-rate component and a lump-
records, depending on the regulatory treatment).
sum payout. Under US terminology, this would
include universal life and whole life.
The distribution mix is available for the largest
35 countries. Channel categories consist of tied
Annuities: individual life-savings products (both
agents, brokers or independent financial advisers,
single and regular premium) that provide a
bancassurance, branches, direct, and others (such
guaranteed credited-rate component and a payout
as retailers and car dealers). These channels are
in the form of an annuity (in other words, a regular
defined later in this section.
monthly payment stream for either a fixed duration
or life). Under US terminology, this would include
The GIP model was built by collecting and analyzing
fixed annuities.
public data (such as national insurance regulators’
data or industry association publications) country
Unit-linked: individual life-savings products
by country and drawing on the insights of our
(both single and regular premium) for which the
global network of local experts. We mapped the
policyholder bears the investment risk and that
local product types and distribution channels to
provide a lump-sum payout. Under US terminology,
the standard of globally accepted definitions.
this would include variable life, variable universal
life, and variable annuities.
Distribution channel definitions
Tied agents work exclusively for one or a few
Group life: group protection, group unit-linked
companies, or for the partners cooperating
products, and group annuities; the largest segment
with a company. Self-employed tied agents are
is corporate pensions.
remunerated on a commission-only basis. Salaried
employed tied agents are remunerated with salary
plus commission.

2019 Global Insurance Pools trends and forecasts: Distribution 9

 Forecasting methodology
Our volumes-forecasting model is based on a series
of historical multivariate regression models that
use both macroeconomic drivers and momentum as
explanatory variables for premium growth.
All our models employ economic forecasts from
Oxford Economics. The macrodrivers we considered
include GDP growth (nominal and real), long-term
and short-term interest rates, penetration, and
equity market returns.

We run panel regressions with random effects
at both country and product-category levels. For
life, P&C, and health, we split countries into two or
three subgroups, based primarily on each country’s
level of economic maturity. We then run separate
regressions for the subgroups at the country and
product-category levels, with particular equation
specifications for each product.
Our global network of local experts reviews the
forecasts produced by our regression models
to adjust for any specificities in local markets
(upcoming regulatory changes, demographic shifts,
or pension or healthcare system reforms).

For our profit-forecasting model we also

developed separate methodologies for life, P&C
and health. For P&C and health, we take a driver-
based approach in which we forecast separately
all components of profit (claims, costs, and
investment income). For each profit component,
we test various specifications, combining
macroeconomic variables (such as GDP growth,
interest rates, and inflation) and time-series
variables (such as momentum effects and mean-
reversion effects). The approach for life was similar.
That said, since life profits are highly sensitive
to capital market and regulatory conditions, any
profit forecast is only valid under the assumption
of stability on both these fronts.

For both P&C and life, we ran panel regressions

grouping similar countries. Overall, the regressions
have generated superior results, with strong r² values,
good stability, and reasonable back-testing behavior.

10 2019 Global Insurance Pools trends and forecasts: Distribution

Authors and acknowledgements

Claudia Max
Associate Partner
Munich

Philipp Klais
Engagement Manager
Munich

Shitij Gupta
Expert
Gurgaon

Nataliya Fedorenko
Specialist
Paris

Varun John Jacob

Analyst
Gurgaon

Kavya Ramjiyani
Analyst
Gurgaon

Tanya Sogani
Analyst
Gurgaon

Shivani Agarwal
Junior Analyst
Gurgaon

The team can be reached at the central email address: global_insurance_pools@mckinsey.com

The authors would like to acknowledge the contributions of colleagues Pierre-Ignace Bernard,
Stephan Binder, Tanguy Catlin, Ari Chester, Bernhard Kotanko, Brad Mendelson, Jörg Mußhoff, and
Kurt Strovink.

Visit McKinsey.com for more on Global Insurance Pools