HSBC - The World in 2030 Report
HSBC - The World in 2030 Report
HSBC - The World in 2030 Report
GLOBAL
September 2018
Disclaimer & Disclosures: This report must be read with the disclosures and the analyst
certifications in the Disclosure appendix, and with the Disclaimer, which forms part of it
ECONOMICS ● GLOBAL
September 2018
Executive Summary
We have refreshed our long-term forecasting framework to make model projections for 75
developed, emerging and frontier economies to assess growth potential and changes in global
rankings by 2030.
Our model focuses on catch-up potential, population (size and shape), human capital
(education and healthcare), politics, openness and technology. Better-educated workers are
more likely to be productive; poorer countries will have room to catch up by simply adopting best
practice and technology available elsewhere; and those with strong governance are more likely
to facilitate investment and growth. Our projections point to the following:
The trend of the past five years, of just below 3% global growth, looks like it could be
sustainable, implying that by 2030, global GDP is about 40% higher than in 2017. Growth in
both EM and DM is projected to be a little weaker than over the past decade but EM now
makes up a larger share of the world.
Over the past decade EM accounted for about half of global growth and on our modelled
estimates, over the coming decade or so, roughly 70% of global growth will be from
countries we currently describe as emerging.
China is set to continue to be the single biggest contributor to global growth over the next
decade and by 2030, will have become the world’s largest economy (see page 11).
One of the most striking rises amongst the rankings will be by India, which is set to become
the world’s third-largest economy in just over a decade, up from seventh today – leap-
frogging the second- and third-largest developed economies of Germany and Japan.
Another five Asian economies feature among our six fastest-growing economies in the
world – Bangladesh, India, Philippines, Pakistan and Vietnam – so that by 2030, the
contribution to global growth from emerging Asia excluding China will be converging on that
of the whole of the group of countries currently classified as developed by MSCI.
There is also continued room for catch-up going beyond 2030. Even in this world, and after
doubling in 2007-2030, average EM GDP per capita is set to remain just a fraction of that in
the west. On our projections it will still be less than 15% of the developed economy average
(roughly 10% today) and China’s will be below 30%.
Demographically, Africa stands out, with its working-age population set to grow by more
than 2.5% per year for the next decade, versus a fall of 0.5% per year in Europe, so that by
2030, Africa will have more people of working age than China.
The small population, demographically challenged, rich economies in Europe slide down
the rankings: Austria and Norway do not even make it into the top-30 by 2030 while
Denmark slips below the top-40.
While poorer countries with younger populations will generally see the sharpest moves up
the rankings, other factors matter too. Improvements in education, healthcare and the rule
of law can still see countries with shrinking working populations hold their position or even
move up the rankings, notably Thailand, Serbia and some of the other CEE countries.
1
ECONOMICS ● GLOBAL
September 2018
In others, technology could offset demographic drag. In our model we have lessened the
impact of shrinking working populations in the countries with the highest number of
robots – Korea, Germany, Singapore and Japan. But others, particularly China, could see
its growth being lifted by this too.
In Latin America, Mexico and Peru stand out, with plenty of room for catch-up, favourable
demographics and relatively robust human capital fundamentals for economies at their
stage of development.
EM countries will account for roughly 50% of global GDP by 2030, which represents a
seismic shift from half that in 2000. As these countries develop and the nature of growth
becomes more domestically oriented and consumer-led, such as we are seeing in China,
the influence on developed markets will rise.
But their impact on other emerging economies, for instance on intra-EM trade and multi-
lateral trade arrangements, will grow too. And as their economic might increases, their
desire for greater political clout in international organisations and suchlike can be expected
to grow too.
Policy challenges
As for policy, the projections and rankings contained in this report are based an assumption that
policymakers will continue to make progress on addressing economic flaws (education, rule of
law etc) and that they avoid wars and remain open to global trade and capital. If these bold
assumptions are wrong, our projections could be wide of the mark.
But there will also be policy challenges and changing priorities that may arise as a consequence
of the projected global shifts in population and economic might – some local, some global.
Environmental challenges will be one of these: it is no coincidence that four of our top-six countries
for projected growth – India, Pakistan, the Philippines and Bangladesh – also top the list of countries
that our ESG analysts have estimated to be the most vulnerable to climate change.
We cannot capture the implications for future growth in our model but environmental considerations,
as well as one of the other most pressing policy challenges of our time – income inequality – are
inevitably leading to renewed discussion about whether GDP itself is any longer the most appropriate
measure for gauging economic growth and well-being (and therefore policy).
2
ECONOMICS ● GLOBAL
September 2018
2018 2030
14%
Brazil 2.1 2.4 Italy
… however average EM GDP per capita still only 14%
of the DM average
2.5
Source: HSBC estimates. Note: GDP figures in constant 2018 USD terms
3
ECONOMICS ● GLOBAL
September 2018
Contents
Executive Summary 1
Fundamental drivers 5
Are EM’s long-term growth
prospects in question? 5
A newer model 6
What do the new country growth
projections look like? 10
What will the global economy look
like in 2030? 12
Disclosure appendix 42
Disclaimer 44
4
ECONOMICS ● GLOBAL
September 2018
Fundamental drivers
The growing risks in many EM countries have been increasingly in focus over the past few
months. Their vulnerability to changing global financial conditions has been exposed by higher
bond yields and a strengthening US dollar. Higher oil prices and growing trade tensions also
threaten additional headaches. For the purposes of this report we set aside these near-term
risks and try to assess the longer-term growth prospects.
This is not the first time we have attempted to take a longer-term view. Back in 2011 in The
World in 2050, 4 Jan 2011, we established a framework for longer-term forecasting and
ultimately used it to make GDP projections for 100 countries in 2050. Demographics, education,
life expectancy, rule of law and other elements of underlying “economic infrastructure” were the
main variables that featured in the model.
6
Australia Indonesia Chile
11 Canada 125 125
Sweden Nigeria
16
Mexico Philippines
21 US 115 115
Russia
26 Japan
France Argentina 105 105
31 Italy Brazil
Spain
36
95 95
36 31 26 21 16 11 6 1
2010 2011 2012 2013 2014 2015 2016 2017 2018
Rank: Forecast GDP growth
Actual Forecast
Source: HSBC World in 2050, IMF Source: HSBC, Thomson Reuters Datastream
To the extent that the accuracy of a 40-year forecast can be assessed after a mere eight years, the
forecasts generated from this model have proved reasonably accurate, at least in terms of the
rankings of growth in a global context: the model correctly projected that the likes of China, India,
Indonesia and the Philippines would outperform their emerging market peers and that developed
market growth would remain much more subdued. The model had more difficulty with commodity
producers, which tend to be less diversified and can be quickly thrown off course by price swings (a
full discussion of how the model has fared is in the appendix on page 27.)
5
ECONOMICS ● GLOBAL
September 2018
6 6
4 4
2 2
0 0
-2 -2
-4 -4
1961 1965 1969 1973 1977 1981 1985 1989 1993 1997 2001 2005 2009 2013 2017
DM EM World
Source: World Bank WDI
This ongoing outperformance by emerging economies means that they have accounted for
more than half of the growth in the global economy since 2010, allowing global growth to remain
remarkably stable in the post-crisis period. Indeed in 2017, the world registered the strongest
year of growth since the immediate bounce-back following the global financial crisis. At 3.1% it
was weaker than in the decade preceding the great recession but the same as the average
pace delivered since 1970 despite weaker demographic drivers in nearly all economies.
We have written at length about how China (see: China and the world, 19 May 2016) and India
(see: India and the world, 12 September 2017) are now crucial to the global growth outlook, but the
rest of the emerging world is playing a part too. One of the biggest questions for economists is how
sustainable is this pace of global growth? And can the emerging world do enough of the heavy
lifting to offset the structural deficiencies of high debt and deteriorating demographics in the west?
A newer model
Young populations with higher growth in working-age population growth will have a greater
share of the population that is of working-age over the next decade or so, helping to lift per-
capita growth rates, not just total growth rates. As these young people age (up to about aged 55
according to UN studies, chart 6), they should become more productive, particularly as
education rates continue to rise across the emerging world.
6
ECONOMICS ● GLOBAL
September 2018
Germany
UK
Absolute productivity US Japan
Turkey
China
Mexico
India
Japan - 1970
Philippines
US - 1970
Kenya
Nigeria China - 1970
0 10 20 30 40 Age 50 60 70 80 90
Source: HSBC, UN population division. Note: Shape of productivity by age - based on academic work using a number of different job types. Shape will vary by country and job-
type so is indicative.
Chart 7 shows how many of the large emerging market economies have a population whose
median age is below 30 and so ageing could have a positive impact on growth potential in
coming years. This is captured in the ‘share of population that is working-age’ variable in the
model, and points to a greater share of the population being employed, paying taxes and
consuming more. Taking the US as a guide, even with their higher household savings rates, the
35-64 age groups have the highest expenditure per person.
7. The world’s young population is in Africa and South & South-East Asia
Median Age
7
ECONOMICS ● GLOBAL
September 2018
8. Some parts of the world have scope for catch-up in terms of technological availability
/100 people Mobile subscriptions per 100 people /100 people
140 140
120 120
100 100
80 80
60 60
40 40
20 20
0 0
1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016
US UK Japan China India Ethiopia
Source: World Bank WDI. Note: Can be >100 due to multiple subscriptions per head. Mobile phone adoption is not a perfect indicator. In some emerging markets still at an
early stage of development, mobile phone ownership is much higher – often because they have no had easy access to landlines – than in some much more advanced
economies. But some of the other benefits of a more technologically-advanced economy on productivity will have been captured in the educational performance and healthcare
indicators in the model.
Our model focuses of six main categories of economic indicators: catch-up potential, population
(size and shape), human capital (education and healthcare), politics, openness and technology:
better educated workers are more likely to be productive; poorer countries will have room to
catch up by simply adopting best practice elsewhere; and those with strong governance are
more likely to facilitate investment and growth.
The full methodology for the model and how we got there is detailed starting on page 32 of the
appendix while full details of the economic infrastructure for the countries covered is overleaf.
We also refresh the list of countries we forecast to include those within MSCI World, Emerging
Market and Frontier indices, as can be seen in table 9. But for the purposes of our groupings into DM
and EM in this report, the EM aggregate includes both “Emerging” and “Frontier” as EM.
9. 75 countries to forecast
________ Developed ________ ________ Emerging ________ ________ Frontier ________
Americas Europe/ Asia Americas Europe/ Asia Europe/ Asia Africa
CEEMEA CEEMEA CEEMEA
Canada Austria Australia Argentina Czech Rep. China Croatia Bangladesh Kenya
US Belgium Hong Kong Brazil Egypt India Estonia Sri Lanka Mauritius
Denmark Japan Chile Greece Indonesia Lithuania Vietnam Morocco
Finland New Zealand Colombia Hungary Korea Kazakhstan Nigeria
France Singapore Mexico Poland Malaysia Romania Tunisia
Germany Peru Qatar Pakistan Serbia Ivory Coast
Ireland Russia Philippines Slovenia Senegal
Israel Saudi Arabia Taiwan Bahrain Burkina Faso
Italy South Africa Thailand Jordan
Netherlands Turkey Kuwait
Norway UAE Lebanon Ghana*
Portugal Oman Ethiopia*
Spain
Sweden Ukraine*
Switzerland
UK
Source: MSCI Country classification. Note: *Countries not in MSCI index but included as covered by HSBC or in Ethiopia’s case have a large population and could play a
significant role in global growth.
8
ECONOMICS ● GLOBAL
September 2018
Human Capital
Gross Primary
Working-age
(children per
Fertility Rate
Change in %
working age
working age
(Trade as %
(2018-2023)
(2018-2023)
(2010 USD)
population
enrolment
Openness
All latest
Share of
woman)
Index**
School
available figures
people
GDP)
unless otherwise
stated
US 55018 0.2 65.4 -1.6 1.9 130 99 3.7 2 26.6
Canada 52815 0.1 66.5 -2.3 1.6 89 101 3.7 1 64.4
Austria 50601 -0.3 66.5 -1.8 1.5 181 102 3.3 1 99.7
Belgium 47078 0.0 64.1 -1.4 1.8 105 103 3.1 1 164.4
Denmark 63158 0.2 63.7 -0.6 1.8 118 102 3.5 1 101.0
Finland 48417 -0.2 61.9 -1.6 1.8 132 100 3.4 1 71.7
France 43675 0.0 62.0 -1.0 2.0 106 107 3.1 1 60.5
Germany 48087 -0.5 65.3 -1.8 1.5 143 102 3.7 1 84.3
Ireland 78448 0.7 64.2 -0.1 2.0 101 101 3.1 1 221.2
Israel 35936 1.4 60.2 -0.2 2.9 125 104 3.7 1 58.4
Italy 35421 -0.5 63.3 -1.1 1.5 164 101 3.1 1 56.2
Netherlands 55648 -0.2 64.6 -1.5 1.7 123 103 3.3 1 153.9
Norway 93943 0.6 65.2 -0.9 1.8 107 100 3.6 1 67.4
Portugal 23581 -0.6 64.7 -0.8 1.2 110 105 2.4 1 78.9
Spain 33389 -0.3 65.7 -0.9 1.4 113 104 2.9 1 62.9
Sweden 59443 0.4 62.3 -0.9 1.9 124 123 3.4 1 83.7
Switzerland 79334 0.2 66.4 -1.7 1.5 131 104 3.7 1 120.4
UK 43397 0.2 63.5 -1.0 1.9 118 102 3.7 1 58.0
Australia 58621 0.7 65.1 -1.4 1.8 114 101 3.5 1 40.0
Hong Kong 39511 -0.8 71.2 -5.1 1.3 261 107 3.2 1 372.6
Japan 49027 -0.7 59.7 -1.1 1.5 141 99 3.5 1 31.2
New Zealand 39070 0.4 64.6 -1.3 2.0 131 99 3.3 1 52.5
Singapore 56108 0.2 71.7 -2.9 1.3 154 101 3.5 4 318.4
Argentina 10649 1.0 64.0 0.2 2.3 151 110 2.9 2 26.3
Brazil 11179 0.6 69.8 -0.2 1.7 102 115 2.7 2 24.6
Chile 15752 0.5 68.4 -0.9 1.8 129 100 3.1 1 56.1
Colombia 7869 0.7 69.0 0.1 1.8 124 114 2.5 3 34.9
Mexico 10132 1.3 66.7 0.6 2.1 92 104 2.7 3 78.1
Peru 6473 1.3 65.6 0.4 2.4 130 103 2.8 2 44.8
Czech Republic 23647 -0.5 65.0 -1.5 1.6 118 99 3.7 1 151.6
Egypt 2988 1.7 61.5 0.1 3.2 106 104 2.5 6 30.0
Greece 23523 -0.4 65.3 -0.5 1.3 112 95 3.0 2 61.6
Hungary 16182 -0.9 66.5 -1.8 1.4 121 102 3.3 3 169.0
Poland 16402 -1.0 67.8 -2.6 1.3 131 110 3.3 1 100.5
Qatar 69573 1.5 84.7 -0.8 1.9 125 104 2.9 6 89.1
Russia 11463 -1.0 67.6 -2.9 1.8 162 102 3.4 7 46.3
Saudi Arabia 21600 1.6 71.7 0.0 2.5 111 116 2.6 7 60.9
South Africa 7716 1.2 65.7 0.4 2.4 123 103 2.7 2 60.4
Turkey 15778 1.0 67.1 0.5 2.0 95 103 2.3 5 46.8
UAE 41886 1.3 84.9 -0.2 1.7 252 111 2.7 7 205.3
China 7849 -0.2 71.2 -1.6 1.6 107 101 2.5 7 37.1
India 2133 1.3 66.4 1.0 2.3 103 115 2.1 2 39.8
Indonesia 4396 1.1 67.5 0.5 2.3 180 103 2.4 2 37.4
Korea 27042 -0.6 72.2 -3.4 1.3 129 98 3.6 2 77.7
Malaysia 12299 1.2 69.4 -0.4 2.0 135 103 3.0 4 128.6
Pakistan 1310 2.1 60.8 0.9 3.4 79 98 1.8 4 25.1
Philippines 3132 1.6 63.6 0.5 2.9 96 113 2.6 3 64.9
Taiwan 22511 -0.8 72.7 -3.6 1.2 129* 98* 3.2 1 120.8
Thailand 6369 -0.3 71.2 -1.4 1.5 221 101 2.7 6 123.1
Croatia 15283 -1.1 65.1 -1.6 1.4 106 95 3.4 1 96.3
Estonia 19707 -0.7 63.7 -1.4 1.7 144 97 3.6 1 154.1
Lithuania 17000 -1.1 65.8 -2.3 1.7 148 101 3.2 1 147.7
Kazakhstan 11351 0.6 64.5 -1.2 2.6 129 108 3.2 7 60.2
Romania 11315 -1.0 66.5 -1.7 1.5 115 89 3.2 2 83.7
Serbia 6167 -0.8 65.8 -1.5 1.6 130 101 3.3 3 107.5
Slovenia 26714 -0.9 65.2 -2.7 1.6 118 99 3.5 1 146.2
Kenya 1263 3.1 57.2 2.0 3.8 82 105 2.2 4 37.9
Mauritius 10606 0.0 70.7 -0.7 1.4 151 102 2.6 1 98.4
Morocco 3433 1.0 65.8 -0.5 2.4 105 110 1.8 5 80.4
Nigeria 2528 2.9 53.4 1.0 5.4 82 94 1.9 3 21.1
Tunisia 4450 0.5 67.7 -1.5 2.1 117 115 2.4 2 90.0
Ghana 1968 2.5 58.3 1.1 3.9 154 105 2.4 1 88.6
Bangladesh 1180 1.6 67.0 1.8 2.1 87 119 2.0 4 38.0
Vietnam 2015 0.5 69.5 -1.4 1.9 125 110 2.6 7 184.7
Sri Lanka 4032 0.2 65.8 -0.2 2.0 141 102 2.9 3 50.5
Source: HSBC, World Bank, IMF, UN Population Division, Penn World Tables, Freedom House International. Note: Sorted by income category and then region. *No data from World
Bank, so Korea taken as a proxy. **Human Capital index is from the Penn World tables and is index of human capital per person, which is related the average years of schooling and the
return to education. Mobile phone adoption may seem high for some emerging markets even compared with developed economies, but the benefits of technology will also be captured in
DM in some of the other human capital indicators.
9
ECONOMICS ● GLOBAL
September 2018
Given the various idiosyncrasies of each country in the global economy, a model to generate
Demographics and catch-up
potential are key longer-term economic forecasts on a broad cross-section of countries will have its limitations.
Some could prove to be off-track in the very near term, should they fall into recession in the
coming year (see What can’t the model capture? on page 15). Nonetheless, we still see value in
making projections out to 2030 in a consistent manner across countries. Our model projections
should certainly give a clear sense of where the economic potential lies: current growth rates
play no role in the projections. In simple terms, countries which have a low starting point in
terms of level of GDP per capita, and have favourable demographics, should grow reasonably
quickly as long as the other components for growth are in place – and the 75 countries in our
sample are shown on chart 11 below.
In practice, our model favours those countries which not only have strong demographics and
…but education, healthcare
and strength of institutions catch-up potential (Ethiopia), but which also have relatively high levels of education, good
matter too health care quality and which are open or have strong political rights. On this basis, our new
model projects that some of the fastest growers will be those in the middle of the top-left
quadrant: India, Bangladesh, Philippines and Vietnam. Some of the others with strong
demographics in the top-left quadrant are held back by the relative weakness of their
institutions: a young rapidly growing working-age population is not supportive for growth if these
young adults are not doing anything productive. Those in the bottom right quadrant have little
scope for catch up and are home to shrinking working-age populations, creating a double
headwind to growth that requires strong human capital or investment in automation to raise
productivity enough to prevent a slowdown in potential growth. Our projections for total GDP
growth by five-year periods out to 2030 for all 75 countries are in the appendix on page 39 but
we show the current rankings for total size of economy and total population on page 11.
10
ECONOMICS ● GLOBAL
September 2018
12. GDP and population rankings in 2018 and model projections for 2030
11
ECONOMICS ● GLOBAL
September 2018
Firstly, these projections suggest that, irrespective of the numerous strains on many emerging
economies, the fundamentals imply that EM-supported global growth of the past decade has
further to run. The trend of the past five years, of just below 3% global growth, looks like it could
be sustainable, implying that by 2030, global GDP is about 40% higher than in 2017.
1965-1970
1970-1975
1975-1980
1980-1985
1985-1990
1990-1995
1995-2000
2000-2005
2005-2008
2012-2017
2018-2023
2023-2028
2028-2030
EM DM Total
Source: World Bank, HSBC estimates. Global GDP growth is calculated using nominal GDP weights rather than PPP weights.
On our projections, both DM and EM growth will be slower in the coming decade than they have
Growing EM contribution
been since the financial crisis but EM now comprises a larger share of the world. Over the past
should allow global growth
to hold up decade EM accounted for about half of global growth and on our modelled estimates, over the
coming decade or so, roughly 70% of global growth will be from countries we currently describe
as emerging. EM growth is projected to be 4.4% in 2018-30 (compared with 4.7% in 2010-17)
and DM growth is seen at 1.5% (down from 1.7%).
80 80
60 60
40 40
20 20
0 0
1960-1965
1965-1970
1970-1975
1975-1980
1980-1985
1985-1990
1990-1995
1995-2000
2000-2005
2005-2008
2012-2017
2023-2028
2028-2030
2018-2023
Unsurprisingly, in 2008-17 the biggest single contribution to global growth came from China. On
Our model suggests five of
our projections China will continue to lead the charge over the next decade or so, but the other
the six fastest-growing
economies will be in Asia notable shift in the composition of global growth will come from the rest of emerging Asia. With
five of our six most rapidly growing economies – Bangladesh, India, Pakistan, Philippines and
Vietnam – all forming part of that region, by 2030 the contribution to global growth by emerging
12
ECONOMICS ● GLOBAL
September 2018
Asia ex China should be converging on that of the whole group of countries currently classified
as developed by MSCI (chart 14).
Like the EM aggregate, China and the rest of Asia may not be set to grow quite as quickly in the
India shows the sharpest
move up the rankings to next decade as in the last, but the region now makes up a bigger share of the world so its
number three contribution to global growth is as high. The most notable shift is set to be in India, with our
projections suggesting it will go from the seventh largest economy in the world to the third by
2030 (table 12).
There is also continued room for catch-up going beyond 2030. Even in this world, and after
doubling between 2007 and 2030, underlying GDP per capita is set to remain just a fraction of
that in the west. On our projections, the EM average will still be less than 15% of the DM
average and China will be below 30%.
For DM it is a mixed picture. In terms of absolute size, the large economy and stronger
demographics in the US keeps it close to the top of the rankings while the sheer size of their
economies means Germany and Japan stay in the top five, despite rapidly ageing populations.
The biggest risers up the rankings are all in Asia. But the small population, demographically
challenged, rich economies in Europe slide down the rankings: Austria and Norway do not even
make it into the top-30 by 2030 while Denmark slips below the top-40.
As for the global population, the shift to EM looks set to continue. India and China, by 2030,
account for 35% of the global population and by 2030, nearly 25% of the world’s working age
population will be in other Asian countries. However, the biggest regional mover will be Africa,
where young, fast-growing populations mean the continent will have more people aged between
16 and 64 than China by 2030.
13
ECONOMICS ● GLOBAL
September 2018
20 20
15 15
10 10
5 5
0 0
North America Europe Latam China Africa India Asia ex-CH, IN
2000 2018 2030 and JP
Source: UN Population Division, HSBC
And while the movements in the share of the global working-age population look quite small
over the next 12 years (chart 16), the divergence between working-age population growth rates
is really striking (chart 17). The shrinkage in China’s working-age population becomes much
more apparent in the latter part of the next decade, while the difference between the American
and European demographic outlook is striking.
17. Demographic divide is clear when it comes to working age population growth rates
% Yr Growth in world working-age (16-64) population % Yr
3.0 3.0
2.5 2.5
2.0 2.0
1.5 1.5
1.0 1.0
0.5 0.5
0.0 0.0
-0.5 -0.5
-1.0 -1.0
North America Europe Latam China Africa India Asia ex-CH, IN
2018-2022 2023-2027 2028-2032 and JP
Source: UN Population Division
14
ECONOMICS ● GLOBAL
September 2018
Known unknowns
There are many ways in which medium- to long-term forecasting models can be wrong. Some are
country-specific: the so-called ‘known unknowns’ out there. The UK, for example, will see its potential
growth shaped by its resulting trade relationship after exiting the EU. Saudi Arabia and other oil
exporters will depend heavily on the oil price itself, but also the pace of any structural reforms to lift
potential growth in non-oil sectors. In Europe, ageing populations could be offset by rising
participation rates for people aged over 65 if pension reforms are successful. All of these are known
risks that are impossible to quantify in a growth model. But there are also a number of other factors
that could significantly alter the entire global outlook, from trade wars to natural disasters:
18. Rising participation for over-65s could 19. The decline in global trade tariffs may
ease decline in working population have come to a halt
% Employment Rate (65+) % % World: Most Favoured Nation weighted average tariff %
12 12 23 23
10 10 21 21
19 19
8 8
17 17
6 6 15 15
13 13
4 4
11 11
2 2 9 9
0 0
7 7
1998 2003 2008 2013 2018 5 5
France Germany Italy UK 1990 1994 1998 2002 2006 2010 2014
Source: Thomson Reuters Datastream Source: WITS
1) Trade wars
The biggest immediate danger to our projections is if the open borders that have delivered so
much prosperity are closed. Recent actions by the US administration are not encouraging on
this front as it is hard to see how such a wave of protectionism could benefit any individual
economy, or the system as a whole. Global growth would inevitably be weaker but as always,
15
ECONOMICS ● GLOBAL
September 2018
there would be distributional effects. The impact on confidence and investment would likely be
negative while the disruption to integrated global supply chains established over the past few
decades would ultimately weigh on living standards. This could be a particular blow to the likes
of Mexico, where our model gives very generous forecasts due to its open economy and
favourable demographic outlook.
At the time of writing the latest round of US tariffs on further USD200bn of imports from China
had just been implemented but our China economists have also written on how such a scenario
could potentially lead to even more rapid intra-EM trade (see: A blessing in disguise: why a
trade war will strengthen China-EM links, 31 August 2018).
2) Technology
We have incorporated a measure of technology usage into our estimates for growth potential.
As explained on page 37, we also lessen the demographic drag in our model projections for the
four countries with the highest number of robots (Korea, Germany, Singapore and Japan) – all
of them face big demographic challenges. Nonetheless, it is highly likely that our model is not
fully capturing the actual impact on GDP that technological advances are already having (for
instance we have not incorporated China’s robotisation plans as discussed on page 16), never
mind the impact that future technologies will have on productivity and well-being.
As recently as 2007, very few would have foreseen the transformational impact that the iPhone and
other smartphones would have on the world in the space of a decade. As we highlighted in Upwardly
mobile: Three themes driving EM growth, 09 October 2017, the arrival of the smartphone in the
hands of young EM consumers could have a transformational impact on access to banking,
educational attainment and healthcare quality, which would lead to faster convergence in this sort of
modelling, meaning that some of the countries with lower quality education and healthcare scores
could see a more rapid catch-up, supporting growth rates in future years.
Austria
Russia
South Korea
Sweden
Netherlands
Italy
France
Portugal
Mexico
Israel
Turkey
Argentina
Estonia
Brazil
Croatia
Indonesia
Singapore
Belgium
Taiwan
Spain
Finland
Poland
Greece
India
Canada
Slovenia
Slovakia
China
Norway
New Zealand
Malaysia
United States
Romania
Philippines
Switzerland
Australia
UK
Hungary
Thailand
Denmark
Czech Rep.
South Africa
Source: IFR
16
ECONOMICS ● GLOBAL
September 2018
21. China’s economy is developing into 22. More automation could offset the
new technologies impact of demographic decline on
productivity
Number of tech 'unicorns' China: Model real GDP growth projections (% Yr)
140 140
120 120 2018-2023
100 100
80 80
60 60 2023-2028
40 40
20 20
2028-2033
0 0
Germany
Korea
Israel
Other
Indonesia
US
UK
India
China
0 1 2 3 4 5 6 7
China with robot offset on demographics China base case
Source: www.CBinsights.com Note: A unicorn company is a start-up with a valuation Source: HSBC estimates. Note: Robot offset implies halving the rate of decline in
over USD1 billion the share of the population that is of working-age.
3) Natural disasters
Natural disasters can send economies seriously off course as their development seeks to
replace what was lost (although they have a temporary upward impact on GDP growth) rather
than make any further leap forward.
4) Migration flows
As discussed on page 29, demographic assumptions tend to be reasonably accurate over the
short term but for many developed markets, population growth is heavily influenced by migration
flows (chart 23) and the sheer volume of new arrivals to Germany in 2015/16 led the UN to
heavily revise its forecasts for the size of the German population over the next few decades
(chart 24). This leaves demographic projections open to error given how susceptible migration
flows are to policy changes from governments. While Germany and Sweden saw that influx,
Singapore has reduced its national quota for foreign workers and Australia is talking about it.
23. Migration drives demographics in DM… 24. …and transforms the demographic
outlook
% Yr Working-age (16-64) population change: % Yr Mn Germany: Total population Mn
2015-2025 84 84
2.0 2.0
1.5 1.5
80 80
1.0 1.0
0.5 0.5 76 76
0.0 0.0
-0.5 -0.5 72 72
-1.0 -1.0
68 68
-1.5 -1.5
-2.0 -2.0 64 64
Australia
Germany
Japan
France
Italy
UK
Sweden
US
Canada
60 60
1950 1970 1990 2010 2030 2050 2070 2090
Natural change Migration Total 2017 update 2015 edition
Source: UN Population Division Source: HSBC, UN Population Division
17
ECONOMICS ● GLOBAL
September 2018
5) Cyclical interruptions
Our model is a structural model of potential supply and therefore ignores cyclical factors and
whether there are ebbs and flows in demand. We do not know when the next downturn will be.
The EM economies most exposed to US tightening and where policymaking has been
questionable, are already set to slide into recession, but this model is not about the cyclical.
It is notable that the two countries which most surprised on the upside in 2011-17 relative to the
projections we set out in the The World in 2050, 4 Jan 2011 were Sweden (where the forecast
was probably too low but rapid debt accumulation no doubt played a role) and Turkey, where
overly loose monetary and fiscal policy contributed to wide imbalances for which it may already
be starting to face the consequences.
18
ECONOMICS ● GLOBAL
September 2018
Energy demand
Projecting a global economic expansion of this scale is one thing. Whether it is feasible in terms
of its impact on global demand for resources is another. Given the increasing energy demands
from the emerging world (chart 25), global power demand could rise dramatically in the coming
years, if recent trends continue (chart 26).
25. On current trends, EM power demand 26. …but with population growth, could
per capita would still be only half of DM lead to a 45% increase in power
by 2030… consumption
kWh/capita Electric power consumption kWh/capita tWh World electric power consumption tWh
10000 10000 40000 40000
35000 35000
8000 8000 30000 30000
25000 25000
6000 6000
20000 20000
4000 4000 15000 15000
10000 10000
2000 2000
5000 5000
0 0 0 0
1970 1980 1990 2000 2010 2020 2030 1970 1980 1990 2000 2010 2020 2030
DM EM DM EM
Source: HSBC, World Bank WDI. Note: Projection assumes average growth rate as Source: HSBC, World Bank WDI, UN population division. Note: Pink and grey bars
past five years are projections based on left hand chart and demographic projections.
The potential challenges are not captured by the model but there can be little doubt that they
are complex. The changing mix of growth over time, particularly in the digital age, means the
relationship between GDP growth and demand for natural resources and labour is by no means
linear. Hence, simple extrapolations are inadequate, especially given the scope for
technological advances to make further rapid progress.
19
ECONOMICS ● GLOBAL
September 2018
Some developments, such as increased customisation and digitisation, should already mean lower
Is growth becoming more or
demand for such resources – think on-demand 3D printing or music streaming (involving lower
less resource intensive?
shipping, storage, commodity and labour costs, as well as lower manufacturing output and less
waste). But other parts of the economy will inevitably increase their call on resources as countries
mature. Technological advance clearly has a role to play in improving energy efficiency and
developing new energy supply – in the way that has already succeeded in considerably lowering the
cost of renewable fuels – but in many countries, particularly in EM, demand for energy is set to grow
much more quickly than GDP if current trends persist. To ensure that the necessary energy
transformations take place will require not just technology but huge investments in infrastructure (see
Re-energising the world: The economics of energy: past, present and future, 8 January 2018).
27. Not all growth requires more natural 28. Demand for renewable energy is rising
resources… fast – albeit from a very low base
% Music sales by format % % Yr Consumption growth by fuel (2016-17) % Yr
100 100 18 18
16 16
80 80 14 14
12 12
60 60 10 10
8 8
40 40 6 6
4 4
20 20 2 2
0 0
0 0
Hydro electric
Oil
Renewables
Nuclear energy
Coal
Natural Gas
2005 2007 2009 2011 2013 2015 2017
Digital Physical
Source: Recording Industry Association of America (RIAA) Source: BP Statistical Review of World Energy June 2018
Air conditioning is a clear example. Various studies have demonstrated the transformational impact
Short-term gains, long-term
that air-conditioning has had on the likes of productivity, healthcare and life expectancy. However, as
pains?
highlighted in a recent edition of The Economist1, the IEA estimates that over the next ten years, one
billion air-conditioners will be installed around the world: more than were put in between 1902 (when
air conditioning was invented) and 2005. The more emissions produced, the more the world warms
and, as incomes rise, the more their use will rise.
At current rates, Saudi Arabia, will be using more energy to
run air-conditioners in 2030 than it currently exports in oil
The Economist, 25 August 2018
The forecasts contained in this report do not attempt to make a claim on the environmental
impact of the projected growth but should readers wish to consider the likely implications of
what are (in some cases) very high growth rates, they should be considered against other
gauges of the sustainability of growth.
For instance, it is notable – if not surprising – that four of our top six countries for projected
Our fastest growing
growth – India, Pakistan, the Philippines and Bangladesh –top a list of 67 countries that our ESG
economies also top the list for
climate change risks analysts2 have estimated to be the most vulnerable to climate change. That ranking is based on an
assessment of both the physical impacts and the associated energy transition risks to gauge which
are better placed to respond to these pressures.
1
Global cooling, The Economist 25 August 2018
2
Fragile Planet: Scoring climate risks around the world, March 2018
20
ECONOMICS ● GLOBAL
September 2018
29. Our projected fast-growers are most at risk from climate impacts
Index Climate vulnerability score Index
8 8
7 High level of 7
6 vulnerability 6
5 5
4 4
3 3
2 2
1 1
0 0
Indonesia (13)
S Africa (43)
Italy (74)
Pakistan (4)
Brazil (28)
China (9)
UK (61)
Poland (34)
USA (57)
UAE (27)
Singapore (46)
Norway (72)
Finland (70)
Bangladesh (1)
Kenya (8)
Chile (32)
S Korea (50)
Canada (71)
Sweden (54)
Thailand (29)
Morocco (25)
Egypt (15)
Malaysia (18)
S Arabia (37)
Australia (60)
Hungary (31)
Japan (75)
Spain (62)
Switzerland (69)
Germany (64)
India (2)
Philippines (5)
Colombia (21)
Mexico (16)
Vietnam (6)
Nigeria (48)
Argentina (24)
Russia (67)
France (68)
Turkey (39)
GDP does not capture this. Some economists3, notably Diane Coyle, have argued that if policy
decisions are to take account of the environmental impact of growth, there should be an assessment
of the extent to which current growth is occurring at the expense of future growth. Hence, the
depreciation of natural resources needs to be accounted for in measures of national income in the
same way as the depreciation of machinery, equipment and infrastructure.
30. Net national income accounts for capital depreciation but not depletion of natural
resources
Mtoe Global carbon dioxide emissions Mtoe
35000 35000
30000 30000
25000 25000
20000 20000
15000 15000
10000 10000
1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
Source: BP Statistical Review
While many governments now have their official statisticians collecting various environmental
indicators, from emissions to air and water quality, we can find no example of a country which is
incorporating it directly into GDP or related measures. Hence all our projections contained in our
long-term growth model are aimed at projecting growth rates on the existing measures of GDP. We
nonetheless recognise that environmental considerations, as well as one of the other most pressing
policy challenges of our time – income inequality – are inevitably leading to renewed discussion
about whether GDP itself is any longer the most appropriate measure for gauging economic growth
and well-being (and therefore policy). This subject forms the basis of the next section.
3
Diane Coyle. GDP: A brief but affectionate history. Princeton University Press 2014
21
ECONOMICS ● GLOBAL
September 2018
Our new projections are for the next decade or so. Based on the last decade, there is a strong
GDP basket to keep
expanding – from prostitution likelihood that at a minimum, what is incorporated in GDP over that period evolves considerably,
to Air B&B which could alter both the size of measured GDP and the growth rate. After all, in the past 15 years,
the standard way in which GDP is measured has altered, mainly through the 2008 changes to the
UN’s system of national accounts and, the case of Europe, from the 2010 version of the European
System of Accounts. The former related to the reclassification of R&D spending to be part of
investment spending and changes to the measurement of financial services. The latter related
primarily to elements of the informal sector, incorporating estimates for the likes of drugs and
prostitution. Each resulted in the overall level of GDP being revised higher than previously reported.
31. The global economy is becoming more 32. …and trade is too
services based…
% GDP EM: Share of GDP % GDP % Total Services as % exports % Total
60 60 40 40
55 55
35 35
50 50
30 30
45 45
40 40 25 25
35 35
20 20
30 30
15 15
25 25
20 20 10 10
1965 1975 1985 1995 2005 2015 1995 1998 2001 2004 2007 2010 2013 2016 2019
Services Industry Eurozone US Japan
Source: Thomson Reuters Datastream Source: Thomson Reuters Datastream
22
ECONOMICS ● GLOBAL
September 2018
Given that every expansion of the GDP measure to date has focused on transactions for which a
Some quality improvements
are being captured… payment is made, it is likely that in the coming years the basket of goods and services included in
GDP will at least be widened to include all services for which money changes hands. So spending on
AirB&B, for instance, could be included going forward.
However, some of these may be captured via greater efficiency and higher productivity. With
Productivity may be higher in
some sectors smart phones now affordable in large parts of even the least developed countries, consumers’
access to banking and other services – or quite simply online information – is an important part
of this; for example, the ability of consumers to compare prices, locate products or even just
access weather reports, or the latest agricultural prices for farmers deciding whether to take
their products to market on a particular day. Increasingly transformational improvements in high-
tech areas relating to healthcare and education can also raise productivity.
Income inequality and well-being may not rise in line with GDP…
Other quality improvements relate more to well-being or “welfare”. Few would argue that GDP is
an adequate measure of well-being. Sure, when GDP is rising and unemployment is low a
greater share of the population should be feeling content but averages can mask a lot of
divergence. Data for the US in charts 33 and 34 highlight the divergence between the mean and
the median incomes, illustrating how the average masks the underlying story. The causes and
issues relating to income inequality, including labour market disruption from technological
advances, are increasingly apparent and have been well documented by ourselves and others.
As we highlighted in The inequality challenge, 18 December 2017, income inequality often goes
hand in hand with a lack of social cohesion and can mean weaker growth.
Furthermore, many of the wealthier countries do not necessarily have the highest quality of life
and are also not without social problems. The US has a higher number of homicides than, say,
Chile. Some of highest average GDP per head countries also have some of the worst statistics
on drug addiction and suicide.
33. Median incomes lag GDP per capita… 34. …and the same applies to wages
1980=100 US 1980=100 1980=100 US 1980=100
200 200 150 150
180 180 140 140
160 160 130 130
140 140 120 120
120 120 110 110
100 100 100 100
80 80 90 90
1980 1985 1990 1995 2000 2005 2010 2015 1980 1985 1990 1995 2000 2005 2010 2015
Real median personal income Real compensation per hour
Real GDP per capita Median weekly real earnings
Source: St Louis Fed Source: St Louis Fed
23
ECONOMICS ● GLOBAL
September 2018
So general well-being and quality of life may well be improving more rapidly than indicated by
GDP, though of course problems arise for governments and households if nominal GDP and
wages do not rise as much as envisaged when governments and households took on their debt.
See the table on page 41 for a summary of environmental, social and governance variables that
measure the progress of countries asides from growth.
Alternative measures
Hence alternative measures of well-being have been devised, such as the Genuine Progress
Indicator (GPI) which is measured in about 20 countries around the world and, in some
cases – including the US – for regions within a country. It incorporates things like volunteer or
household work and subtracts for the negative effects of income inequality and environmental
damage as well as crime.
The fact is it is hard to measure well-being, whereas it is quite straightforward to measure GDP
No easy substitute for GDP
in a timely way and make cross-country comparisons. Along with population and average
income levels for household groups, it is the best indication of market size. Besides, for the
most part, the Social Progress Index from the Social Progress Initiative shows there is a clear
correlation between incomes and social progress (chart 35). Growth in income and an
improvement in social progress do not always go hand in hand though. Over the past three
years, Nigeria has seen the biggest improvement in its social progress index, thanks to
improvements in information access and personal freedoms, but income growth, especially on a
per-capita basis, has been extremely disappointing. At the other end of the spectrum, Japan
has seen greater improvements in terms of social progress than Canada and Australia, despite
weaker income growth.
24
ECONOMICS ● GLOBAL
September 2018
Greece
75 Tunisia Peru Brazil
Mexico Kuwait
Jordan Malaysia
70 Saudi Arabia
Ukraine Thailand South Africa Turkey
Philippines Sri Lanka Russia
65 Egypt
Kazakhstan
Ghana Lebanon
60 Morocco
Senegal Indonesia China
India Lower level of
55 social progress for
Bangladesh Kenya level of income per
Pakistan capita
50 Burkina Faso Nigeria
45 Ethiopia
40
6 7 8 9 10 11 12
GDP per capita (2018, natural log)
Source: IMF, World Bank, Social Progress Initiative
Policy
The projections and rankings contained in this report are based on countries’ potential to catch
up with more developed nations. We assume that policymakers will continue to make progress
on addressing economic flaws (education, rule of law etc) and that they avoid wars and remain
open to global trade and capital. Some of our bold assumptions may turn out to be overly
optimistic for some countries while others may over-deliver in ways that allow them to overcome
any structural headwinds. Below we focus more on the policy challenges and priorities that may
arise as a consequence of the projected global shifts in population and economic might – some
global, some local.
We project that EM countries will account for roughly 50% of the world by 2030, which represents a
EM will desire greater
political clout on the world
seismic shift from half that in 2000. As these countries develop and the nature of growth becomes
stage… more domestically oriented and consumer-led, such as we are seeing in China, the influence on
developed markets will rise. But their impact on other emerging economies, for instance on intra-EM
trade and multi-lateral trade arrangements, will grow too. And as their economic might increases,
their desire for greater political clout in international organisations and suchlike will grow too.
But even in some of these emerging economies the policy priorities for their own economies will start
…but their changing growth
priorities will impact DMs to shift, though the pace at which they do so will likely depend on their stage of development.
as well We discussed the environmental implications of high growth rates in the most rapidly growing
countries on page 21. Already, China has made clear its desire for cleaner, fairer, more
innovative growth (including an ambition to be world leader in artificial intelligence by 2030).
This will have implications for growth elsewhere, not least as a consequence of China’s
changing import mix. Unlike much of the past 15 years when China’s public investment floated
all boats, particularly commodity producers, its growth mix in the coming decade can be
expected to give a bigger lift to exporters of industrial and office machinery and even consumer
25
ECONOMICS ● GLOBAL
September 2018
goods and services. Germany, Japan, Korea and Taiwan are likely to be among the bigger
winners on the machinery side while big service providers of tourism, education and financial
services will also gain. (See page 36 for a discussion of the specific adjustments we have made
to our projections for Australia and New Zealand.) Within a decade, the growth priorities of other
emerging economies that currently lag China could shift too.
For many DM governments, a backdrop of weaker growth relative to the past few decades will
Demographic challenges in
DM means retirement and present its own challenges. While much of the slowdown is demographic and per capita growth is
pension policies will have projected to hold up, lower aggregate GDP growth implies lower revenues for heavily indebted
to adjust governments in the west to sustain increasing age-related spending. The pressure to raise official
retirement ages is set to intensify. Within the next decade some countries will start to see the impact
of policies already in the pipeline. In the UK, the state pension age will rise to 66 years from 2020, to
67 around 2026, and then be linked to life expectancy. In Spain, reforms approved in 2011 will
increase the retirement age from 65 years to 67 by 2027, and similar moves are underway in Greece
and Portugal. The increase should be welcomed but the move is very small and retirement age
changes take a long time to have any effect on public finances and participation rates.
Allowing more immigration could help to ease the demographic strains but the past few years have
Immigration to remain an
opportunity and a challenge also demonstrated that higher immigration can cause political strain. Some countries are already
resorting to controlled immigration schemes focusing on specific skills but the numbers who want to
migrate from their native countries could actually rise. Despite stronger growth in EM, average GDP
per head in EM will still only be about 14% of the DM average by 2030, as we note on page 13. But
for the middle income parts of the population, the means to migrate will rise. Various studies4 by the
OECD and others using UN data have shown that African emigration rates to OECD countries are
strongly related to GDP per capita.
Moreover, with the overall economy not growing as quickly as in the past and the likely ongoing
Weaker growth in DM means
addressing income inequality disruption to labour markets from robotisation and automation, many DM governments’ near-term
will remain a priority priorities will be to address income inequality. They are well aware of the dangers – politically and
economically – that divergences in income growth pose. But for many governments, which have
accumulated huge debts since the global financial crisis, this involves big political choices (see:
The inequality challenge, 18 December 2017). Do they facilitate the conditions for the market to
deliver more inclusive growth? Or is the only way to curb the anger stemming from income
inequality to embark on much more redistribution or government borrowing? Or retreat into
protectionism? Only time will tell.
4
See: Coppel, Jonathan, Jean-Christophe Dumont, and Ignazio Visco. 2001. “Trends in Immigration and
Economic Consequences, OECD Economic Department Working Paper no. 284 and An Age-old Question,
September 2016
26
ECONOMICS ● GLOBAL
September 2018
Back in January 2011, we published The World in 2050, 4 Jan 2011. Now, nearly eight years on,
we revisit our model and forecasts made to assess where we were right, where we were wrong
and where we can improve our thinking.
To start, we can look at where our previous long-run estimates failed to pick up a given
trajectory in terms of growth. Some forecasts were clearly too optimistic – Italy, Spain, Brazil,
Argentina and Russia stand out – but in some cases the projections were too pessimistic: many
of the faster-growing EM countries outperformed our assumptions as did some developed
markets for which our model estimated very weak trend growth assumptions – meaning that
they were not hard to beat, such as Sweden, the US and even Japan.
Brazil
Germany
Korea
Nigeria
Colombia
Indonesia
Italy
France
Switzerland
Argentina
Mexico
Poland
Turkey
India
Norway
New Zealand
Singapore
Chile
Malaysia
Philippines
Spain
UK
US
Canada
Sweden
Australia
Thailand
China
South Africa
DM EM
Delivered 2010-17 World in 2050 projections
Source: HSBC World in 2050 forecasts, IMF
But why did we go wrong? Looking at table 37, we draw out where the largest absolute errors
have been across the major countries that we cover. The main reasons for substantial error are
politics or commodity prices – with those countries with reasonably high projections seeing their
performance hampered by the fall in prices in 2014, but also the fact that the biggest commodity
producers tend to be less diversified probably meant the model over-estimated their growth
potential. Politics has also played a role in Argentina, Brazil and Turkey, while the eurozone
crisis made it very hard for Spain and Italy to achieve anywhere near the growth rates we
forecast in 2011 over the whole period – although it is notable that Spain has grown at roughly
2.4% per year since 2012, not far from our trend assumptions.
27
ECONOMICS ● GLOBAL
September 2018
37. Most of the reasons for the biggest errors would not have been forecast by the model
World in 2050
Delivered growth projections Error Rationale
2010-17 2012-17
Brazil 0.4 -0.5 3.3 -2.9 Commodity/domestic shock
Russia 1.4 0.2 4.2 -2.8 Commodity/domestic shock
Chile 3.2 2.2 5.9 -2.7 Commodity
Argentina 1.2 0.7 3.4 -2.2 Domestic factors
Philippines 6.2 6.6 8.4 -2.2 Forecast for pace of catch up overly optimistic
Spain 0.8 1.9 2.8 -2.0 Eurozone crisis, but better since 2012
Sweden 2.3 2.8 0.4 1.9 Model penalised it for already high GDP per capita
Malaysia 5.2 5.2 7.1 -1.9 Very strong forecast due to low start point
Italy -0.1 0.3 1.4 -1.5 Eurozone crisis but also forecast too optimistic
Turkey 6.6 6.1 5.4 1.2 Stimulus, creating imbalances
Source: HSBC World in 2050 forecasts, IMF
Some countries have done better, mainly those in EM that outperformed assumptions that are
likely constrained by econometric modelling: China, India and Indonesia have all outperformed
by around 1ppt over the past seven years. But the biggest two upside surprises came in
Sweden (where the forecast was very low and debt has built up) and Turkey, where continued
economic stimulus has led to wide imbalances for which it may already have started to feel the
effects]. This underlines one of the key points about long-term forecasts: they attempt to project
a long-term potential growth rate, not to forecast every twist and turn in the economy, which can
be heavily influenced by fiscal and monetary policy and which in turn may mean an economy
diverges markedly from its long-term trend for a period.
5 Indonesia Philippines
Singapore Malaysia
Nigeria Colombia
Poland
10
Chile
New Zealand Thailand
Rank: Delivered GDP growth
Australia Korea
15
Sweden Canada Mexico
US
20 UK
South Africa Germany
Norway
Switzerland
25 Russia
France Argentina
Japan
Spain
30 Brazil
Italy
35
35 30 25 20 15 10 5 0
Rank: Forecast GDP growth
Source: HSBC, IMF World Economic Outlook
28
ECONOMICS ● GLOBAL
September 2018
Some parts of the model have already proven to be accurate: our projections for Poland,
France, Canada and Germany were on track while the UK and South Africa have encountered
shocks in terms of the Brexit vote and commodity prices/politics respectively but the model
ended up being reasonably accurate: long-term forecasting can be right for the wrong reasons.
But, tellingly, the model did a very good job of getting the rankings right. Growth can be under or
over-estimated by a model for a number of reasons that shock the global economy in a positive
or negative way, but getting the rankings right is a good judge of whether a model is picking up
the right country differentials when it comes to trend growth.
The model also uses variables which are hard to extrapolate. Inflation and the share of GDP
that is government spending do not necessarily correlate with development; thus forecasting
future periods becomes more spurious. Equally, running panel regression tests for some of the
variables within the original model suggests alternative measures for education, healthcare and
the political environment are more statistically significant and can be updated.
Some variables will clearly be important for determining economic growth: human capital,
physical capital, innovation and strong institutions. Some make logical sense: better educated
workers are more likely to be productive, poorer countries will have room to catch up and those
with strong governance will facilitate investment and growth.
39. Demographic trends can weigh on 40. …and the shape of the population
growth… matters too
% Yr Developed Markets (5 year annualised) % Yr ppts Japan (5 year annualised) % Yr
1.2 4.5 0.6 6
1.0 4.0 0.4 5
0.8 3.5
0.2 4
0.6 3.0
0.0 3
0.4 2.5
0.2 2.0 -0.2 2
0.0 1.5 -0.4 1
-0.2 1.0 -0.6 0
-0.4 0.5
-0.8 -1
-0.6 0.0 1980 1990 2000 2010 2020
1980 1990 2000 2010 2020 Change in %population that is working age (LHS)
Working-age population (LHS) GDP (RHS) GDP (RHS)
Source: HSBC, World Bank, UN Population Division Source: HSBC, World Bank, UN Population division
Populations
Demographics are important to economic forecasts. Size of population is crucial for
demand – particularly the working-age population, which is more likely to produce and consume
goods and services. It seems logical to think that a larger share of the population that is of
working-age would be conducive to faster growth rates too – as the shape of the population will
become more productive as well as putting less strain on the fiscal position of a government.
We use indicators for both the growth rate of the population and the shape of the population for
29
ECONOMICS ● GLOBAL
September 2018
this reason. Demographic indicators have a lot of use in this form of modelling because we can
use the forecast for the next five-year period. There are risks around these forecasts as there
can be shifts in fertility rates, migration and life expectancy (see: Are demographics destiny? 30
July 2018) but demographic assumptions are reasonably accurate over the short term.
Human capital
It is all well and good having a fast-growing population, but if the potential output per head does
not change, this may not be fully realised in headline growth rates. Across the world education
levels are rising very quickly (chart 42), and historically have been shown to be a good indicator
of future growth potential. While many long-run forecasts use the educational data from Robert
Barro and Jong-Wha Lee5, this data is only available at five-year intervals. We instead use two
sources: the World Bank’s data on primary enrolment: the share of the population undertaking
primary education, a good measure of the breadth of educational attainment; and we also use
the broader ‘human capital’ index from the Penn World Tables, which accounts for educational
attainment as well as enrolment. Both are shown to be more statistically significant when
modelling than the Barro-Lee set.
41. Better human capital can spur growth 42. …and education quality is on the rise
in EM… in the poorest countries
% Yr EM growth by human capital quartile % Yr WEF Quality of Primary School education:
3.0 3.0 Score 1 (worst) - 7 (best)
5.0 5.0
2.5 2.5
4.0 4.0
2.0 2.0
3.0 3.0
1.5 1.5
2.0 2.0
1.0 1.0
1.0 1.0
0.5 0.5
0.0 0.0
0.0 0.0 Bangladesh Kenya Philippines India
Bottom 25% 25-50% 50-75% Top 25% 2013 2018
GDP per capita growth, five years forward
Source: Penn World Tables, HSBC calculations. Note: Taking countries with a Source: HSBC, World Economic Forum Global Competitiveness Index
human capital index score that are in each percentile, we then took a 5-year forward
annualised GDP growth and the chart shows a simple average.
Capital quality is a difficult item to incorporate into projections of potential growth rates.
Underdeveloped capital stock creates room for future growth, but equally may mean there is not
enough capital to spur future growth. For this reason, many variables are statistically
insignificant: the capital stock level as a share of GDP (or population), investment as a share of
GDP and the cost of investment all failed to show meaningful predictive power. In fact, the
5
Barro, Robert and Jong-Wha Lee, 2013, "A New Data Set of Educational Attainment in the World, 1950-
2010." Journal of Development Economics, vol 104, pp.184-198.
30
ECONOMICS ● GLOBAL
September 2018
share of GDP that is investment appears to be more coincident than leading, as faster growth
leads to a more rapid pick-up in investment relative to other parts of the economy – hence why
many academic models may find the variable significant.
43. Poorer countries typically grow more 44. Investment correlates with GDP growth
quickly in real time, not leads
% Yr Average five year forward GDP per capita % Yr % Yr US % GDP
growth by income quintile 5 25
3.5 3.5
4
3.0 3.0 23
3
2.5 2.5
2 21
2.0 2.0
1.5 1.5 1
19
1.0 1.0 0
0.5 0.5 -1 17
1960 1970 1980 1990 2000 2010
0.0 0.0 5-year GDP growth (LHS)
0-20% 20-40% 40-60% 60-80% 80-100% Investment % GDP (RHS)
Source: World Bank WDI, HSBC calculations. Note: uses past 25 years of data Source: World Bank WDI, HSBC
across the World Bank data available.
Institutions
The quality of institutions will play a big role in delivering on the potential growth in a country.
Without a regulatory environment that makes investment attractive and facilitates investment,
potential growth will be lower. Quantitatively, this creates challenges because the nature of such
things makes them hard to quantify and many of the widely used indicators in the field have a
reasonably short history. The Freedom House indices6 have a longer time series, back to the
1970s, providing a history for political rights, civil liberties and whether an economy is free or
not. We find political rights to be the best catch-all in this category and to be significant, even
though there can be sizeable differences between countries that are given the same score: e.g
Singapore and Pakistan currently both score 4 (on a range of 1 to 7).
6
Freedom House, Freedom in the world index.
31
ECONOMICS ● GLOBAL
September 2018
Technology/productivity
The quality of technology is a key driver in many countries as a multiplier on human and
physical capital stock. But finding a variable to model is tricky. The Penn World Tables provide a
Total Factor Productivity (TFP) index for each country, but this is more of a residual than an
input. We could look at the rollout of internet usage, but much of the development has been
very recent or not a relevant indicator (broadband lines, for example, have stopped growing in
EM countries due to mobile leapfrogging). We have used mobile phone adoption (as this starts
earlier) as a proxy for technological development, which is found to be statistically significant.
Indicators such as the share of GDP spent on R&D, while ideal, have too short a time-series to
be appropriate for this sort of modelling.
Mobile phone adoption is not a perfect indicator. In some emerging markets still at an early
stage of development, mobile phone ownership is much higher – often because populations
have no had easy access to landlines – than in some much more advanced economies. But
some of the other benefits of a more technologically advanced economy on productivity will
have been captured in educational performance and healthcare indicators.
47. Some parts of the world have scope for catch-up in terms of technological availability
/100 people Mobile subscriptions per 100 people /100 people
140 140
120 120
100 100
80 80
60 60
40 40
20 20
0 0
1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016
US UK Japan China India Ethiopia
Source: World Bank WDI. Note: Can be >100 due to multiple subscriptions per head.
Openness
Historically, trade openness has been a strong driver of economic growth across the world.
Academic texts such as that from Romain Wacziarg and Karen Horn Welch in 20087, which
finds trade liberalisation contributing to a roughly 1.5ppts higher annual growth rate. However, in
recent months, the idea that openness is conducive to growth is up for question, with trade war
risks weighing on the economic growth opportunities of many countries.
We have to pick a dependent variable first – and given we are trying to estimate a trend growth
rate, five years forward seems appropriate. Much longer, and new factors will influence growth,
and much shorter would make our projections even more vulnerable to shocks. It is also telling
that the IMF World Economic Outlook forecasts five years out.
7Wacziarg, R., and Welch, K.H., 2008. Trade Liberalization and Growth: New Evidence, The World Bank
Economic Review, vol. 22, no. 2, pp. 187–231
32
ECONOMICS ● GLOBAL
September 2018
Eurozone
Argentina
Greece
South Korea
Germany
Russia
Portugal
Japan
Italy
Turkey
Brazil
Colombia
Indonesia
Ireland
France
Saudi Arabia
Norway
Czech Republic
Poland
Mexico
New Zealand
Singapore
Malaysia
Chile
India
Hong Kong
Switzerland
Sweden
Australia
Canada
UK
US
Spain
Hungary
China
Thailand
South Africa
Household Corporate
Source: BIS
From this stage of a rough list of variables to consider, we then have to think about data availability.
Indicators such as private sector debt/GDP would be excellent indicators to include, but a long-term
history does not exist for many countries. The same applies to technology indicators where access to
smartphones or 3G would cut our sample of countries so short (in terms of the number of years we
look at) that we would be unable to run a meaningful regression.
These indicators will have some correlation between them: for example, health-based indicators
such as infant mortality, life expectancy and fertility rates will be correlated, not just with each
other but with measures of income and development. This must be accounted for when running
regressions and thinking about how inter-connected the variables are within the data set.
From this set of indicators we built a panel of data that covers as broad a range of countries and
time periods as possible. We want to include countries with a range of income levels too, given
the forecasting model will be looking to forecast developed and emerging markets. The panel
contains the countries below.
33
ECONOMICS ● GLOBAL
September 2018
The next discussion is over what estimation technique to use within the regression. We used a
panel data set, and a panel regression8.
Many of the independent variables in table 49 were thrown out by being deemed insignificant in
determining future trend growth. Some may have proved useful, but the lack of data availability
meant the number of panel observations was too low. Many indicators, however, remained in the
model and we were left with the variables in table 52 below. For four of the variables, GDP per
capita, share of population of working age, mobile ownership and openness, we take the natural
logarithm in order to put less weight on extremities. Much of the impact will be seen through the initial
improvement in these variables, with the change in GDP per capita from USD10,000 to USD20,000
likely to have a greater impact than a move from USD80,000 to USD90,000 on the potential pace of
growth. This equation has an R2 to future GDP growth of 0.46.
For subsequent time periods, we use the UN’s projections for demographic indicators coupled with
an “improvement” in the other indicators based on their pace of GDP growth in the previous period.
8While using fixed effects for cross sections may have appeal for this sort of model, the output gave
extremely unrealistic outcomes for the model. Instead, we used dummy variables for the countries the
model struggles with.
34
ECONOMICS ● GLOBAL
September 2018
R-squared 0.460
Source: HSBC. Note: The Human Capital Index may have a p-stat of 0.22, but improves the predictive power of the model with its inclusion.
Alternative models
While our final regression uses 13 explanatory variables (three of which are dummies), many
other options were tried. We used the shares of the economy as the building blocks as well as
trialling a human plus physical capital approach. The challenge within a lot of these variables is
that they either have too short a time series (many are only available from 1991 onwards), are
not available for a large selection of countries, or simply are not significant when put into a
regression as a leading indicator for growth.
53. The model was not too bad when faced with out-of-sample data – errors were shocks
% Yr Model vs actual GDP growth: 2012-17 % Yr
8.0 8.0
7.0 7.0
6.0 6.0
5.0 5.0
4.0 4.0
3.0 3.0
2.0 2.0
1.0 1.0
0.0 0.0
Russia
Japan
Korea
Indonesia
Brazil
Germany
Italy
France
Saudi Arabia
Colombia
Switzerland
Turkey
Poland
Mexico
Argentina
India
Singapore
New Zealand
US
Canada
Malaysia
Philippines
Norway
Spain
UK
Sweden
Chile
China
Australia
Thailand
Hong Kong
South Africa
Model Actual
Source: HSBC
This model also got some countries wrong. Our forecasts for Saudi Arabia were far too
optimistic, but given the 4% pa rise in working-age population and a 3ppt increase in the share
of the population that is working age over the period, it is hard to come up with a more benign
forecast using this sort of modelling when the demographic change within the period is so
dramatic –an outlier compared with the rest of the world.
35
ECONOMICS ● GLOBAL
September 2018
But, encouragingly, the other large errors are more understandable. Brazil, Russia and
Argentina would have been hard to forecast in this type of model given that political and
commodity shocks are out of the remit of this sort of modelling. A few outperformed, with
Sweden the other stand-out surprise, where the fast growth was partly due to big changes in the
near-term demographic picture with the large number of migrant arrivals in 2015 and 2016.
As discussed on page 15, there are many ways in which medium-term forecasting models can
be wrong. While some are the so-called ‘known unknowns’ we mentioned like Brexit and oil
price swings, there are some countries with which the model struggles for what we consider
reasonably straightforward reasons and which in many ways represent a limitation of a model
that cannot capture the impact of changes in in other countries or even global factors. Some of
these we can make specific adjustments for:
Australia and New Zealand, whose economies are supported by both migration and overseas
visitors, are hard to forecast accurately. Both economies are wealthy and although they have
favourable demographics by developed market standards, forecasting tourism is difficult,
especially when it has been one of the key drivers of growth in recent years. Since 2012, tourist
arrivals to both countries have been rising more than 8% per year, with Chinese arrivals rising
by 18% per year over the same time frame – and helping to support the domestic economies
(see: Downunder Digest: Services exports and the AUD, 30 March 2016). Given that in the
1990s Chinese arrivals accounted for less than 1% of the total, the changing dynamic is hard to
pick up in a model that looks at historical data.
The same could apply to Thailand, as Jingyang Chen highlighted in a recent note (Thailand
tourism: A story of resilience, 15 August 2018). But, on top of this, both Australia and New
Zealand have continually seen increases in immigration numbers. Knowing the number of
consumers and possibly producers will be higher than base-line figures, we nudge up the
demographic growth rates for both countries.
54. Tourism is driving growth in some 55. …with arrivals from new sources
parts of the world…
000s Australia % 000s New Zealand %
900 18 350 14
800 16 300 12
700 14
250 10
600 12
500 10 200 8
400 8 150 6
300 6
100 4
200 4
100 2 50 2
0 0 0 0
1991 1995 1999 2003 2007 2011 2015 2019 1991 1995 1999 2003 2007 2011 2015 2019
Tourist Arrivals (LHS) Tourist Arrivals (LHS)
Share from China (12m MAV, RHS) Share from China (12m MAV, RHS)
Source: Thomson Reuters Datastream Source: Thomson Reuters Datastream
36
ECONOMICS ● GLOBAL
September 2018
Another country where the model appears to be giving a forecast that is slightly “out” is South Africa;
where the model gave us a roughly 3.5% growth rate over the next decade, which seems too high
given its recent history. South Africa is a good example of where limiting factors on its own growth
rate are not statistically significant across a broad range of countries. This includes youth
unemployment (ie, a rapidly-growing population of working age but many not engaged in productive
activity) – a huge headwind to growth in South Africa with nearly 60% of 18-24 year-olds
unemployed. However, when putting the indicator into a cross-country model, it is either found to be
insignificant or to have a positive beta, which is intuitively wrong. To account for this, we adjust the
share of the population that is of working age lower, due to the extreme disconnect between the two
variables for South Africa. Our ‘adjusted’ model projected rate for South Africa is 2.0% for the
average period to 2030, which is still higher than our actual forecasts for the next couple of years, as
GDP growth is held back by rising taxes and reform paralysis.
56. Youth unemployment in South Africa is 57. …meaning we need to adjust the
the highest in the world… demographics indicators
% Youth (16-24) unemployment rate, 2017 % 80
60 60 75
Brazil
India
Saudi Arabia
Italy
US
UK
Spain
South Africa
Africa
35
70 75 80 85 90 95 100
Share of population > 15
Source: World Bank WDI. Note: ILO measure for comparability Source: HSBC, World Bank WDI, UN Population Division.
58. Korea’s demographics are 59. …and the population shape is set to
concerning… weaken sharply
% Yr Korea: Annual change in population 2017-27 % Yr Korea: Population by age-group
4.0 4.0 100% 100%
37
ECONOMICS ● GLOBAL
September 2018
As our colleague James Lee has written, this shift in the Korean economy towards higher-tech
sectors that have lower employment intensity (ie higher labour productivity) has allowed growth
to hold up (see: Korea outlook: Implications of a flourishing high-tech industry, 23 January
2018). Based on data from IFR, there are four countries (Korea, Germany, Singapore, Japan)
with a particularly high number of robots, who also have a noticeable demographic challenge.
We can therefore lower demographic drag to account for this, as the impact on GDP growth
should be reduced – with Korea by a bigger factor than the other three.
Austria
Russia
Netherlands
France
Portugal
Mexico
Israel
South Korea
Sweden
Italy
Turkey
Argentina
Estonia
Brazil
Croatia
Indonesia
Singapore
Belgium
Taiwan
Spain
Finland
Greece
India
Canada
Slovenia
Slovakia
China
Norway
New Zealand
Poland
Malaysia
United States
Romania
Philippines
Switzerland
Australia
UK
Hungary
Thailand
Denmark
Czech Rep.
South Africa
Source: IFR
So these adjustments lead us to the projections over the page. We also include a table of the
comments from our country economists – who are responsible for our two-year ahead central
GDP forecasts published in our global and regional quarterly economics reports – about where
they see the potential risks to our long-term global model-based projections, on page 40.
38
ECONOMICS ● GLOBAL
September 2018
61. HSBC long-term growth model projections for real GDP growth
Trend growth pace (to 2030) 2018-2023 2023-2028 2028-2033
1 Bangladesh 7.1 7.3 7.0 7.2
2 India 6.2 6.1 6.1 6.5
3 Ethiopia 5.8 5.4 5.9 6.6
4 Pakistan 5.7 5.2 5.9 6.5
5 Philippines 5.7 5.4 5.7 6.2
6 Vietnam 5.6 5.4 5.7 6.2
7 Ghana 5.4 4.9 5.6 6.0
8 Kenya 5.3 5.1 5.4 5.6
9 China 5.2 5.3 5.4 4.7
10 Jordan 5.0 4.8 5.0 5.3
11 Burkina Faso 5.0 4.5 5.1 5.7
12 Senegal 4.9 4.3 5.2 5.6
13 Indonesia 4.8 4.7 4.8 4.9
14 Peru 4.6 4.5 4.6 4.8
15 Egypt 4.6 3.8 5.0 5.6
16 Mexico 4.6 4.5 4.6 4.7
17 Sri Lanka 4.2 4.3 4.2 4.0
18 Malaysia 4.2 4.1 4.1 4.6
19 Serbia 3.8 3.5 3.9 4.2
20 Bahrain 3.7 4.4 3.3 3.2
21 Colombia 3.7 3.9 3.6 3.4
22 Tunisia 3.5 3.0 3.8 4.3
23 Argentina 3.4 3.3 3.5 3.7
24 Morocco 3.3 3.0 3.4 3.8
25 Cote d'Ivoire 3.3 2.8 3.4 4.0
26 UAE 3.2 3.4 3.2 2.7
27 Brazil 3.2 3.3 3.1 3.1
28 Thailand 3.1 3.6 3.0 2.4
29 Oman 3.1 3.0 3.0 3.4
30 Czech Republic 3.1 2.7 3.3 3.4
31 Hungary 3.1 2.6 3.4 3.4
32 Chile 3.0 3.1 2.9 2.7
33 Kazakhstan 2.9 1.9 3.4 4.1
34 Poland 2.9 2.3 3.2 3.6
35 Estonia 2.8 2.6 2.9 3.1
36 Romania 2.8 2.3 3.4 2.6
37 Saudi Arabia 2.7 2.6 2.6 2.9
38 Israel 2.6 2.1 2.9 3.1
39 Turkey 2.6 2.9 2.5 2.5
40 Ukraine 2.6 2.2 2.7 3.2
41 Croatia 2.5 2.4 2.6 2.8
42 Ireland 2.5 2.4 2.6 2.3
43 South Africa 2.4 2.2 2.5 2.7
44 Taiwan 2.3 2.6 2.1 2.1
45 Hong Kong 2.3 2.4 2.1 2.6
46 Lebanon 2.2 2.9 1.6 1.8
47 Singapore 2.2 2.5 2.0 1.9
48 Kuwait 2.2 1.9 2.5 2.0
49 Nigeria 2.2 1.7 2.4 2.9
50 Lithuania 2.1 2.1 1.9 2.5
51 Korea 2.1 2.4 1.9 1.7
52 Mauritius 2.0 2.7 1.6 1.4
53 Portugal 2.0 2.4 1.9 1.4
54 Greece 1.9 2.3 1.9 0.9
55 Sweden 1.8 1.8 1.9 1.9
56 Slovenia 1.8 1.5 2.0 2.1
57 Belgium 1.8 1.9 1.7 1.8
58 US 1.8 1.8 1.6 2.0
59 Qatar 1.8 1.9 1.8 1.3
60 New Zealand 1.7 1.8 1.7 1.8
61 Australia 1.7 1.5 1.7 2.0
62 UK 1.5 1.6 1.5 1.4
63 Spain 1.5 2.0 1.3 0.6
64 Germany 1.3 1.7 1.1 0.9
65 Denmark 1.3 1.7 1.2 0.7
66 Netherlands 1.2 1.6 1.0 0.8
67 Russia 1.2 0.5 1.3 2.6
68 France 1.2 1.3 1.2 1.1
69 Switzerland 1.1 1.4 0.9 0.8
70 Finland 1.1 0.9 1.1 1.4
71 Canada 1.1 1.0 0.9 1.5
72 Norway 1.0 1.1 0.9 0.8
73 Austria 0.9 1.4 0.6 0.4
74 Italy 0.9 1.4 0.7 -0.1
75 Japan 0.6 0.6 0.7 0.3
Source: HSBC. Note that these projections are the projections from our long-term global model. They may differ from the forecasts from HSBC’s country
economists , the details of which can be found in the Global Economics Quarterly and the regional quarterlies.
39
ECONOMICS ● GLOBAL
September 2018
62. Countries where our economists see some risks to our long-term projections
Country Comment
Australia/New Favourable demographics, driven by migration could help to keep the average age of the population lower and spur growth
Zealand higher - particularly if infrastructure and housing investment is improved.
Chile The central bank sees trend growth closer to 3.5% and they raise an interesting point which is the potential for labour force
changes. In this case, either from immigration or larger participation of the elderly or female population. They see those
scenarios potentially bringing higher growth.
China Reform will be a key determining factor of China’s medium to long term growth outlook. Deepening market-oriented reforms in
areas including state owned enterprises, industrial policy, financial sector and labour markets can substantially lift the country's
total factor productivity whilst resolving the debt problem.
Colombia To reach 3.7% we would need a catch-up in terms of infrastructure or major education reform to increase the share of the
population getting college education over the next decade.
India India has undertaken several reforms over the last few years, namely the Goods and Services Tax reform, a new Insolvency
and Bankruptcy Code, steps to promote digitization, steps to clean up the real estate sector, and making the RBI an Inflation
Targeting central bank. As growth dividends from these trickle in, potential growth could be higher than that identified in the
model for the next few years.
Indonesia We think long-run growth can be slightly higher, due to incremental improvements in capital spending and better macro stability,
especially in terms of managing inflation and more sensible spending on both hard and soft infrastructure.
Korea Korea’s trend growth is likely to slow, but the slowdown may partly be offset by capital investment and automation, which has
been leading to productivity growth.
Mexico Although the 4.6% may seem a bit ambitious, there are strong arguments to support it. We should eventually see a big leap in
terms of growth, especially considering demographics and the openness of the economy. However, some of the things that have
capped growth (quality of institutions for example) are hard to measure and Mexico's openness leaves it exposed to a slowdown
in trade.
Morocco, Jordan, High youth unemployment poses a risk to the economies’ ability to harness their demographic potential. In Jordan's case - as
Tunisia, Egypt long as the political problems on its border persist, it is difficult to imagine Jordan taking growth this high, whatever its potential
might be. The growing proportion of the educated population choosing to work overseas may maintain growth through
remittances, but their departure will discourage domestic productive investment, providing a further cap to growth.
Nigeria Given its demographics and natural resources and the impact that even a modest improvement in structural policymaking could
achieve, there is every prospect that growth exceeds the rate identified by the model.
Oman & Bahrain Maturing conventional oil reserves pose a risk to long-term outlook, especially if the two economies do not succeed in reforming
their fiscal accounts and diversifying their economic base.
Peru Peru's story so far has been catching up to other countries, and has relied on mining, so may be harder to continue over a longer
time frame.
Philippines In recent years the government has significantly ramped up spending on fixed investment after years of chronic underspending,
which has increased potential growth. Due to sufficient fiscal space and recent tax reforms, this spending can be sustained, and
there seems to be a credible trend of improved economic management over the course of the past three administrations, thus
transcending politics. Coupled with the long-term demographic boom and relatively strong levels of tertiary education, a higher
rate of long-run growth is very possible in the Philippines.
Saudi Arabia The success of the Kingdom’s ambitious reform programme, or lack thereof, could shift Saudi Arabia’s growth trend in either
direction.
Turkey The risk to the long-term trend growth assumption of 2.6% for Turkey lies to the upside. The government’s assessment of
potential growth is around 5%, which makes it likely that domestic policy is likely to be loosened if growth remains below that
perceived potential for a few years.
UAE Maintaining strong demographic growth will be contingent on the country’s ability to attract expatriate labour as it has done in
recent decades.
Source: HSBC
40
ECONOMICS ● GLOBAL
September 2018
Total (recorded+unrecorded)
Sensitivity to extreme events
100,000 people)
GINI coefficient
consumption
Argentina 6.2 4.6 5.3 42.4 47.7 18.8 13.6 6.5 9.8 117 3.9
Australia 5.7 2.3 3.1 34.7 59.8 12.6 30.1 1.0 10.6 14 5.3
Austria 5.7 6.0 6.8 30.5 55.9 11.3 12.1 0.5 11.6 22 5.7
Bangladesh 4.5 1.2 5.2 32.4 33.2 11.4 8.4 2.5 0.0 177 2.9
Belgium 4.5 5.2 4.8 27.7 47.8 20.1 6.7 1.9 12.1 52 5.4
Brazil 6.0 3.5 5.1 51.3 52.7 30.2 19.5 26.7 7.8 125 4.1
Canada 8.2 4.9 5.3 34.0 61.5 11.6 21.7 1.7 8.9 18 5.7
Chile 7.5 2.8 6.6 47.7 48.7 17.2 16.2 3.6 9.3 55 4.8
China 6.5 2.0 5.1 42.2 63.7 - 8.8 0.7 7.2 78 4.7
Colombia 6.1 2.2 3.1 50.8 57.7 17.5 15.5 26.5 5.8 59 3.8
Croatia 6.3 6.1 5.7 30.8 45.4 31.3 32.9 0.9 8.9 51 4.6
Czech Rep. 3.9 3.3 6.4 25.9 51.9 10.5 21.3 0.7 14.4 30 4.6
Denmark 4.6 8.4 6.6 28.2 59.4 12.0 20.9 1.0 10.4 3 5.5
Egypt 3.1 7.1 4.4 31.8 23.0 30.8 4.5 3.2 0.4 128 4.1
Estonia 7.2 8.3 6.8 32.7 56.4 13.4 10.1 3.2 11.6 12 5.1
Finland 7.5 8.5 5.9 27.1 55.0 20.1 1.4 1.6 10.7 13 5.4
France 4.8 3.2 7.1 32.7 51.6 24.6 37.1 1.6 12.6 31 6.1
Germany 5.6 5.7 6.0 31.7 55.6 7.1 12.9 0.8 13.4 20 6.0
Ghana - - - 42.4 65.5 15.2 7.1 1.7 2.7 120 3.3
Greece 5.6 4.5 4.2 36.0 45.0 47.3 22.9 0.8 10.4 67 4.9
Hong Kong - - - - 54.8 9.8 21.4 0.3 - 5 6.7
Hungary 6.2 3.6 6.7 30.4 48.0 12.9 14.6 1.5 11.4 48 4.4
India 3.8 1.7 4.3 35.1 23.4 10.1 14.1 3.2 5.7 100 4.2
Indonesia 5.1 3.8 5.0 39.5 52.8 13.4 17.4 0.5 0.8 72 4.5
Ireland 6.8 7.5 6.0 31.9 53.2 17.2 20.9 0.6 13.0 17 5.1
Israel 2.2 3.9 4.2 41.4 59.3 7.3 18.2 1.4 3.8 54 5.4
Italy 5.2 6.3 6.0 34.7 40.5 37.8 44.6 0.8 7.5 46 5.4
Japan 5.9 5.0 4.4 32.1 50.3 5.1 34.0 0.3 8.0 34 6.3
Kazakhstan 6.3 6.0 3.0 26.9 66.7 3.9 2.8 4.8 7.7 36 4.2
Kenya 2.9 3.2 6.6 48.5 76.3 21.3 5.4 5.8 3.4 80 3.5
Korea 5.3 6.5 5.0 31.6 52.6 10.4 16.3 0.7 10.2 4 6.1
Lithuania 6.8 6.7 5.4 37.4 55.7 14.4 13.4 6.0 15.0 16 4.7
Malaysia 5.1 4.7 3.0 46.3 54.3 10.5 11.5 1.9 0.9 24 5.5
Mauritius 4.0 6.7 4.8 35.8 45.5 23.9 20.4 2.7 3.6 25 4.8
Mexico 4.4 3.1 4.4 43.4 43.0 6.9 14.4 16.3 6.5 49 4.3
Morocco 4.0 4.7 5.1 40.7 25.2 19.9 24.9 1.0 0.6 69 4.4
Netherlands 5.6 7.8 3.6 29.3 58.7 10.8 12.0 0.6 8.7 32 6.4
New Zealand 7.1 5.7 6.8 - 65.4 12.7 29.0 0.9 10.7 1 5.5
Nigeria 3.0 5.3 5.1 43.0 51.8 12.4 5.4 9.8 13.4 145 2.0
Norway 7.1 8.0 4.7 27.5 61.4 10.4 6.2 0.6 7.5 8 5.0
Pakistan 3.6 0.9 6.4 30.7 24.2 6.6 10.4 7.8 0.3 147 3.0
Peru 6.4 3.9 4.3 43.8 62.5 15.2 7.7 7.2 6.3 58 3.8
Philippines 3.6 0.4 6.2 40.1 48.0 7.7 8.9 9.8 6.6 113 3.4
Poland 5.4 3.6 5.2 31.8 48.3 17.7 31.0 0.7 11.6 27 4.7
Portugal 6.0 4.5 5.2 35.5 54.1 23.9 30.2 1.0 12.3 29 5.6
Qatar 2.1 8.5 2.7 - 99.3 0.5 11.7 8.1 2.0 83 5.8
Romania 6.3 4.0 6.2 28.3 44.4 20.6 26.5 1.5 12.6 45 3.8
Russia 7.8 3.7 3.7 37.7 63.8 16.1 29.1 11.3 11.7 35 4.9
Saudi Arabia 2.8 5.5 2.9 - 22.2 24.2 8.8 1.5 0.2 92 5.2
Serbia 6.1 2.4 5.4 28.5 45.4 34.9 28.1 1.1 11.1 43 4.1
Singapore 3.6 8.5 4.2 - 60.4 9.1 8.5 0.2 2.0 2 6.5
Slovenia 5.8 3.5 6.2 25.4 52.0 15.2 29.5 1.2 12.6 37 4.8
South Africa 3.8 4.4 3.9 63.0 48.8 53.5 10.2 34.3 9.3 82 4.3
Spain 5.9 6.8 6.2 36.2 52.6 38.6 58.8 0.7 10.0 28 5.9
Sri Lanka 4.0 1.0 6.4 39.8 35.9 21.6 18.6 2.9 4.3 111 3.8
Sweden 7.2 8.4 6.4 29.2 62.0 17.9 16.4 1.1 9.2 10 5.6
Switzerland 5.3 6.9 7.2 32.5 62.6 8.1 42.5 0.7 11.5 33 6.3
Thailand 5.4 0.8 5.4 37.8 60.2 3.7 12.0 3.5 8.3 26 4.7
Tunisia 4.1 7.1 4.1 35.8 26.1 34.7 20.5 3.1 1.9 88 3.8
Turkey 4.5 7.0 5.6 41.9 33.5 20.5 17.7 4.3 2.0 60 4.5
UAE 2.6 8.5 3.3 - 47.5 12.1 12.4 0.7 3.8 21 6.3
UK 5.0 3.7 5.4 33.2 57.5 13.0 25.1 0.9 11.4 7 6.0
United States 6.8 2.2 4.6 41.5 57.0 9.2 32.7 4.9 9.8 6 6.0
Vietnam 5.1 1.3 5.5 34.8 72.0 7.2 3.5 1.5 8.3 68 3.9
Source: HSBC (A new metropolis), World Bank, WEF. Note: Dark grey is a top score within each metric, dark red is a low score.
41
ECONOMICS ● GLOBAL
September 2018
Disclosure appendix
Analyst Certification
The following analyst(s), economist(s), or strategist(s) who is(are) primarily responsible for this report, including any analyst(s)
whose name(s) appear(s) as author of an individual section or sections of the report and any analyst(s) named as the covering
analyst(s) of a subsidiary company in a sum-of-the-parts valuation certifies(y) that the opinion(s) on the subject security(ies) or
issuer(s), any views or forecasts expressed in the section(s) of which such individual(s) is(are) named as author(s), and any other
views or forecasts expressed herein, including any views expressed on the back page of the research report, accurately reflect
their personal view(s) and that no part of their compensation was, is or will be directly or indirectly related to the specific
recommendation(s) or views contained in this research report: Janet Henry and James Pomeroy
Important disclosures
This document has been prepared and is being distributed by the Research Department of HSBC and is intended solely for the
clients of HSBC and is not for publication to other persons, whether through the press or by other means.
This document is for information purposes only and it should not be regarded as an offer to sell or as a solicitation of an offer to
buy the securities or other investment products mentioned in it and/or to participate in any trading strategy. Advice in this document
is general and should not be construed as personal advice, given it has been prepared without taking account of the objectives,
financial situation or needs of any particular investor. Accordingly, investors should, before acting on the advice, consider the
appropriateness of the advice, having regard to their objectives, financial situation and needs. If necessary, seek professional
investment and tax advice.
Certain investment products mentioned in this document may not be eligible for sale in some states or countries, and they may
not be suitable for all types of investors. Investors should consult with their HSBC representative regarding the suitability of the
investment products mentioned in this document and take into account their specific investment objectives, financial situation or
particular needs before making a commitment to purchase investment products.
The value of and the income produced by the investment products mentioned in this document may fluctuate, so that an investor may
get back less than originally invested. Certain high-volatility investments can be subject to sudden and large falls in value that could
equal or exceed the amount invested. Value and income from investment products may be adversely affected by exchange rates,
interest rates, or other factors. Past performance of a particular investment product is not indicative of future results.
HSBC and its affiliates will from time to time sell to and buy from customers the securities/instruments, both equity and debt
(including derivatives) of companies covered in HSBC Research on a principal or agency basis.
Analysts, economists, and strategists are paid in part by reference to the profitability of HSBC which includes investment banking,
sales & trading, and principal trading revenues.
Whether, or in what time frame, an update of this analysis will be published is not determined in advance.
For disclosures in respect of any company mentioned in this report, please see the most recently published report on that company
available at www.hsbcnet.com/research. HSBC Private Banking clients should contact their Relationship Manager for queries
regarding other research reports. In order to find out more about the proprietary models used to produce this report, please contact
the authoring analyst.
42
ECONOMICS ● GLOBAL
September 2018
Additional disclosures
1 This report is dated as at 25 September 2018.
2 All market data included in this report are dated as at close 20 September 2018, unless a different date and/or a specific
time of day is indicated in the report.
3 HSBC has procedures in place to identify and manage any potential conflicts of interest that arise in connection with its
Research business. HSBC's analysts and its other staff who are involved in the preparation and dissemination of
Research operate and have a management reporting line independent of HSBC's Investment Banking business.
Information Barrier procedures are in place between the Investment Banking, Principal Trading, and Research businesses
to ensure that any confidential and/or price sensitive information is handled in an appropriate manner.
4 You are not permitted to use, for reference, any data in this document for the purpose of (i) determining the interest
payable, or other sums due, under loan agreements or under other financial contracts or instruments, (ii) determining the
price at which a financial instrument may be bought or sold or traded or redeemed, or the value of a financial instrument,
and/or (iii) measuring the performance of a financial instrument.
43
ECONOMICS ● GLOBAL
September 2018
Disclaimer
Legal entities as at 30 November 2017 Issuer of report
‘UAE’ HSBC Bank Middle East Limited, Dubai; ‘HK’ The Hongkong and Shanghai Banking Corporation Limited, Hong Kong; HSBC Bank plc
‘TW’ HSBC Securities (Taiwan) Corporation Limited; 'CA' HSBC Securities (Canada) Inc.; HSBC Bank, Paris Branch; HSBC 8 Canada Square, London
France; ‘DE’ HSBC Trinkaus & Burkhardt AG, Düsseldorf; 000 HSBC Bank (RR), Moscow; ‘IN’ HSBC Securities and Capital E14 5HQ, United Kingdom
Markets (India) Private Limited, Mumbai; ‘JP’ HSBC Securities (Japan) Limited, Tokyo; ‘EG’ HSBC Securities Egypt SAE, Telephone: +44 20 7991 8888
Cairo; ‘CN’ HSBC Investment Bank Asia Limited, Beijing Representative Office; The Hongkong and Shanghai Banking Fax: +44 20 7992 4880
Corporation Limited, Singapore Branch; The Hongkong and Shanghai Banking Corporation Limited, Seoul Securities Website: www.research.hsbc.com
Branch; The Hongkong and Shanghai Banking Corporation Limited, Seoul Branch; HSBC Securities (South Africa) (Pty) Ltd,
Johannesburg; HSBC Bank plc, London, Madrid, Milan, Stockholm, Tel Aviv; ‘US’ HSBC Securities (USA) Inc, New York;
HSBC Yatirim Menkul Degerler AS, Istanbul; HSBC México, SA, Institución de Banca Múltiple, Grupo Financiero HSBC;
HSBC Bank Australia Limited; HSBC Bank Argentina SA; HSBC Saudi Arabia Limited; The Hongkong and Shanghai Banking
Corporation Limited, New Zealand Branch incorporated in Hong Kong SAR; The Hongkong and Shanghai Banking
Corporation Limited, Bangkok Branch; PT Bank HSBC Indonesia; HSBC Qianhai Securities Limited
This document is issued and approved in the United Kingdom by HSBC Bank plc for the information of its Clients (as defined in the Rules of FCA) and those of its affiliates only. If this research is
received by a customer of an affiliate of HSBC, its provision to the recipient is subject to the terms of business in place between the recipient and such affiliate. In Australia, this publication has
been distributed by The Hongkong and Shanghai Banking Corporation Limited (ABN 65 117 925 970, AFSL 301737) for the general information of its “wholesale” customers (as defined in the
Corporations Act 2001). Where distributed to retail customers, this research is distributed by HSBC Bank Australia Limited (ABN 48 006 434 162, AFSL No. 232595). These respective entities
make no representations that the products or services mentioned in this document are available to persons in Australia or are necessarily suitable for any particular person or appropriate in
accordance with local law. No consideration has been given to the particular investment objectives, financial situation or particular needs of any recipient.
The document is distributed in Hong Kong by The Hongkong and Shanghai Banking Corporation Limited and in Japan by HSBC Securi ties (Japan) Limited. Each of the companies listed above
(the “Participating Companies”) is a member of the HSBC Group of Companies, any member of which may trade for its own account as Principal, may have underwritten an issue within the last
36 months or, together with its Directors, officers and employees, may have a long or short position in securities or instruments or in any related instrument mentioned in the document. Brokerage
or fees may be earned by the Participating Companies or persons associated with them in respect of any business transacted by them in all or any of the securities or instruments referred to in
this document. In Korea, this publication is distributed by either The Hongkong and Shanghai Banking Corporation Limited, Seoul Securities Branch ("HBAP SLS") or The Hongkong and Shanghai
Banking Corporation Limited, Seoul Branch ("HBAP SEL") for the general information of professional investors specified in Article 9 of the Financial Investment Services and Capital Markets Act
(“FSCMA”). This publication is not a prospectus as defined in the FSCMA. It may not be further distributed in whole or in part for any purpose. Both HBAP SLS and HBAP SEL are regulated by
the Financial Services Commission and the Financial Supervisory Service of Korea. This publication is distributed in New Zealand by The Hongkong and Shanghai Banking Corporation Limited,
New Zealand Branch incorporated in Hong Kong SAR.
The information in this document is derived from sources the Participating Companies believe to be reliable but which have not been independently verified. The Participating Companies make no
guarantee of its accuracy and completeness and are not responsible for errors of transmission of factual or analytical data, nor shall the Participating Companies be liable for damages arising out
of any person’s reliance upon this information. All charts and graphs are from publicly available sources or proprietary data. The opinions in this document constitute the present judgement of the
Participating Companies, which is subject to change without notice. From time to time research analysts conduct site visits of covered issuers. HSBC policies prohibit research analysts from
accepting payment or reimbursement for travel expenses from the issuer for such visits.
This document is neither an offer to sell, purchase or subscribe for any investment nor a solicitation of such an offer. HSBC Securities (USA) Inc. accepts responsibility for the content of this
research report prepared by its non-US foreign affiliate. All US persons receiving and/or accessing this report and intending to effect transactions in any security discussed herein should do so
with HSBC Securities (USA) Inc. in the United States and not with its non-US foreign affiliate, the issuer of this report. In Singapore, this publication is distributed by The Hongkong and Shanghai
Banking Corporation Limited, Singapore Branch for the general information of institutional investors or other persons specified in Sections 274 and 304 of the Securities and Futures Act (Chapter
289) (“SFA”) and accredited investors and other persons in accordance with the conditions specified in Sections 275 and 305 of the SFA. This publication is not a prospectus as defined in the
SFA. It may not be further distributed in whole or in part for any purpose. The Hongkong and Shanghai Banking Corporation Limited Singapore Branch is regulated by the Monetary Authority of
Singapore. Recipients in Singapore should contact a "Hongkong and Shanghai Banking Corporation Limited, Singapore Branch" representative in respect of any matters arising from, or in
connection with this report. HSBC México, S.A., Institución de Banca Múltiple, Grupo Financiero HSBC is authorized and regulated by Secretaría de Hacienda y Crédito Público and Comisión
Nacional Bancaria y de Valores (CNBV).
The document is intended to be distributed in its entirety. Unless governing law permits otherwise, you must contact a HSBC Group member in your home jurisdiction if you wish to use HSBC
Group services in effecting a transaction in any investment mentioned in this document. HSBC Bank plc is registered in England No 14259, is authorised by the Prudential Regulation Authority
and regulated by the Financial Conduct Authority and the Prudential Regulation Authority and is a member of the London Stock Exchange. (070905)
In Canada, this document has been distributed by HSBC Securities (Canada) Inc. (member IIROC), and/or its affiliates. The information contained herein is under no circumstances to be construed
as investment advice in any province or territory of Canada and is not tailored to the needs of the recipient. No securities commission or similar regulatory authority in Canada has reviewed or in
any way passed judgment upon these materials, the information contained herein or the merits of the securities described herein, and any representation to the contrary is an offense.
If you are an HSBC Private Banking (“PB”) customer with approval for receipt of relevant research publications by an applicable HSBC legal entity, you are eligible to receive this publication. To
be eligible to receive such publications, you must have agreed to the applicable HSBC entity’s terms and conditions (“KRC Terms”) for access to the KRC, and the terms and conditions of any
other internet banking service offered by that HSBC entity through which you will access research publications using the KRC. Distribution of this publication is the sole responsibility of the HSBC
entity with whom you have agreed the KRC Terms.
If you do not meet the aforementioned eligibility requirements please disregard this publication and, if you are a customer of PB, please notify your Relationship Manager. Receipt of research
publications is strictly subject to the KRC Terms, which can be found at https://research.privatebank.hsbc.com/ – we draw your attention also to the provisions contained in the Important Notes
section therein.
© Copyright 2018, HSBC Bank plc, ALL RIGHTS RESERVED. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, on any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise, without the prior written permission of HSBC Bank plc. MCI (P) 116/01/2018, MCI (P) 016/02/2018
[1104632]
44
Issuer of report:
HSBC Bank plc
8 Canada Square
London, E14 5HQ, United Kingdom
Telephone: +44 20 7991 8888
Fax: +44 20 7992 4880
Website: www.research.hsbc.com
Main contributors
Janet Henry
Global Chief Economist
HSBC Bank plc
+44 20 7991 6711 | janet.henry@hsbcib.com
Janet Henry was appointed as HSBC’s Global Chief Economist in August 2015. She was previously HSBC’s Chief European Economist and
is a member of Handelsblatt’s Shadow ECB Council. Janet joined HSBC in 1996 in Hong Kong where she worked as an Asian economist
in the run-up to, and aftermath of, the Asian crisis. From 1999 to 2007 she was a Global Economist during which time her original work
on globalisation – including the global determinants of local inflation – was widely recognised. Janet’s career began at the Economist
Intelligence Unit where she worked as an Asian economist in London and Hong Kong.
James Pomeroy
Economist
HSBC Bank plc
+44 20 7991 6714 | james.pomeroy@hsbc.com
James is a global economist at HSBC. He joined the Economics team in 2013 having previously worked within the Asset Allocation research
team. His global work focuses on longer-term trends and themes, and the impact that they have on the economy and policy decisions today.
Demographic data is at the heart of much of his work, but he has also written about urbanisation, the role of technology in the economy
and how the world is moving away from cash. Alongside this, he provides economics coverage of Scandinavia. James holds a BSc in
Economics from the University of Bath.