Download report - Standard Chartered Bank

l Global Research l
Special Report – Economics
Money supply – The forgotten indicator
Highlights
 Theory and historical evidence suggest that broad money and GDP
growth should be closely linked. But the link has weakened in the
face of financial innovation. We examine an alternative measure of
money, Divisia, and believe it can resurrect money as an indicator.
 Standard money supply data simply sums the components included.
Divisia money weights components by their usefulness in transactions.
We find support for studies that show Divisia money growth to be a
good indicator of the stance of monetary policy and a useful predictor
of GDP growth. Our Divisia measures are also based on broad
measures of money which has a bigger impact on economic activity.
 We provide our own in-house estimates of Divisia money for China,
the euro area, India and Japan. For China we produce our own
estimates of broad money supply, M3 and M4. Our global index also
includes the US and UK. We plan to provide regular updates on all
six major economies.
Enam Ahmed +44 20 7885 7735
[email protected]
Senior Economist, Thematic Research
Standard Chartered Bank
Samantha Amerasinghe +44 20 7885 6625
[email protected]
Economist, Thematic Research
Standard Chartered Bank
Chidu Narayanan +65 6596 7004
[email protected]
Economist, Asia
Standard Chartered Bank, Singapore Branch
Acknowledgements
We would like to acknowledge the contribution of
John Calverley, Chief Economist,
Calverley Economic Advisors Ltd.
 Divisia money growth is higher than two to three years ago, except
in the UK; but it remains slower than pre-2008, except in Japan,
suggesting that monetary conditions are still not buoyant. We
believe this supports our view that the global upswing will continue
but stay disappointing and that further stimulus will be necessary.
Important disclosures can be found in the Disclosures Appendix
All rights reserved. Standard Chartered Bank 2016
https://research.sc.com
Special Report – Economics: Money supply – The forgotten indicator
Contents
Infographic
3
Executive summary
5
Why look at Divisia money?
5
What Divisia money growth says
5
Conclusion – Add Divisia money to your dashboard
6
Why money supply is important
8
Theory and history
8
Money supply and quantitative easing
9
How Divisia money is calculated
10
Why Divisia money is superior
12
Country studies
14
Global – Picking up but still lacklustre
14
China – Fastest rate in three years
15
Euro area – A strong bounce in 2015
16
India – Running slowly since 2012
17
Japan – Up with Abenomics
18
UK – Slower recently
19
US – Stuck in low gear
20
Why don’t central banks use Divisia?
22
Conclusion – Add Divisia to your dashboard
22
Appendix 1: The quantity theory of money
25
Appendix 2: Calculating Divisia money
27
Practical problems in calculating Divisia money
Global Research Team
10 March 2016
30
32
2
Special Report – Economics: Money supply – The forgotten indicator
Infograph ic
DIVISIA MONEY TRENDS
Latest rates, %
Japan
Change in last 2 years
Change since 2007
3.5
US
4.0
UK
6.8
Euro area
7.2
India
11.8
China
15.4
0
4
8
12
16
-4
-2
0
2
4
-12
-8
-4
0
4
OUR TAKE
Japan
Slow but higher than pre -crisis
US
Slow but up since 2013; supports our below -consensus
growth forecast
UK
Moderate, but down since 2013
Euro
area
Strong, but much higher since 2013; growth surprise?
India
Moderate, well down on pre - crisis
China
Moderate, up since 2013; supports our above -consensus
growth forecast
Source: Standard Chartered Research
10 March 2016
3
Special Report – Economics: Money supply – The forgotten indicator
CALCULATING DIVISIA MONEY
Cash
Checking accounts
Time deposits
M2
30
10
10
10
Adjusted for usefulness in transactions*, then…
Cash
Checking accounts
Time deposits
½
Divisia
M2
10
10
25
5
*Based on a benchmark rate of 4% and a time deposit rate of 2%
WHY DIVISIA IS BETTER
It captures the relative ‘moneyness’ of each component
It can better accommodate new instruments
It is a theoretically consistent approach
It provides a more reliable signal of GDP trends
Source: Standard Chartered Research
10 March 2016
4
Special Report – Economics: Money supply – The forgotten indicator
Executive summary
Why look at Divisia money?
Divisia money uses a sophisticated
weighting mechanism
Economic theory suggests that growth in broad money supply is a good gauge of
monetary policy stance and likely GDP trends. In practice, financial innovation has
made ‘simple sum’ (SSM) measures of money unreliable as an indicator in recent
years. Divisia (named after French economist Francois Divisia, 1889-1964) money
uses a sophisticated weighting mechanism to adjust for the relative ‘moneyness’ of
different components, so it can take account of substitution between them, and the
emergence of new instruments. This makes it a more trustworthy money indicator
than official series. We find it performs much better as an indicator, and did so
especially during the 2008-09 global financial crisis (GFC).
We present data unavailable elsewhere
In this report we present Divisia money growth estimates for six major countries.
Currently neither the US nor China publish a money indicator broader than M2, but it
is important to look at broad measures to gauge liquidity properly. Our data covers
M4 for China, India, the UK and the US, and M3 for the euro area and Japan (though
definitions vary). We use the official Divisia series for the UK, the only country to
publish regular estimates, while US data is calculated by the Center for Financial
Stability. The estimates for China, the euro area, India and Japan are our own.
What Divisia money growth says
Money growth has been weak since 2008
Divisia money has been weak
despite QE
Our measure of the global Divisia broad money supply (aggregating the six major
countries using GDP weights) illustrates the weakness of money growth since the
GFC, despite large-scale quantitative easing (QE). While central banks have been
adding money by purchasing securities, private-sector deleveraging has subtracted
money. Moreover, not all central bank purchases impact the broad money supply;
purchases from banks may only lead to higher central bank reserves.
Money is picking up, but is still lack-lustre
Divisia money growth picked up in 2015, led by acceleration in the euro area and
China. But the global measure, as well as the country measures (except Japan), are
still growing more slowly than pre-2008 (Figures 1-2). This supports our view that the
global upswing will continue but remain disappointing and more stimulus is needed.
Figure 1: Global Divisia money has been weak since 2008
Divisia and simple sum money, % y/y
10
Figure 2: Divisia money is turning up except in Japan and
the euro area, % y/y
30
25
8
SSM
20
CN
15
6
DM
4
IN
10
EA
UK
5
JP
0
2
-5
0
2007
2008
2009
2010
Source: Standard Chartered Research
10 March 2016
2011
2012
2013
2014
2015
-10
2007
US
2008
2009
2010
2011
2012
2013
2014
2015
Source: Standard Chartered Research
5
Special Report – Economics: Money supply – The forgotten indicator
China – The fastest rate in three years
Divisia is not a crystal ball, but an
indicator worth watching
Our measure of Divisia M4 has picked up over the last year to its fastest rate in three
years (Figure 3, page 7). We believe this pick-up reflects monetary easing, including
cuts in interest rates and reserve requirements as well as easing in macro-prudential
measures (see Special Report, 23 November, 2015, ‘Asian housing: Ride the cycle’).
This momentum supports our view that an imminent hard landing is unlikely and our
above-consensus forecast of 6.8% GDP growth for 2016.
Euro area – A strong bounce in 2015
Since late 2014 Divisia M3 growth has accelerated markedly, reaching 7.2% y/y
(Figure 4). This surge puts money growth only slightly below the 2003-07 average
and likely reflects both active QE and the recent rise in lending to the private sector.
But GDP growth has been lacklustre and most indicators suggest it will remain
modest: Our 2016 GDP forecast of 1.4% is close to consensus. In contrast, strong
Divisia money growth could be pointing to a growth surprise.
India – Running slowly since 2012
Trends in Divisia M4 fit well with the growth slowdown and the fall in the inflation rate
since 2012 (Figure 5). Divisia M4 has firmed slightly over the last year, suggesting
economic growth could pick up or at least continue at recent moderate rates.
Japan – Up with Abenomics
Growth in Divisia M3 has picked up since QE was expanded in 2013 as part of
‘Abenomics’ (Figure 6). But money growth is still only modest, suggesting QE is
struggling to boost broad money. That said, our analysis finds the relationship
between money and GDP in Japan is not as reliable as in other countries.
UK – Slower recently
In the UK, Divisia growth is a good match with GDP trends, much better than the
simple-sum M4 measure. Currently Divisia M4 is growing at 6.8%, below the 7-10%
growth rate seen pre-2008 and also down from the level in 2013 (Figure 7). GDP
growth has indeed slowed since 2013 from 3% to about 2%. This is consistent with
the Bank of England (BoE) remaining on hold, in our view.
US – Stuck in low gear
Our analysis finds that Divisia money fits better than official M2 with trends in GDP.
US Divisia M4 is expanding at 4% currently, well below the 6-7% rate prevailing
before the GFC and helping to explain why GDP growth remains disappointing
(Figure 8). Slow Divisia money growth supports our below-consensus 2016 GDP
forecast of only 1%, with GDP growth held down also by demand leaking abroad due
to the strong US dollar.
Conclusion – Add Divisia money to your dashboard
Our analysis, as well as academic studies, finds that Divisia money is a good
measure of monetary policy stance and a useful indicator for GDP trend. Divisia
money is not a magic crystal ball, but we believe analysts should add it to their
dashboard. It usually performs better than simple-sum measures. Divisia measures
will likely become more important in emerging countries as financial systems deepen,
widening the gap with simple sum measures. We plan to provide regular updates.
10 March 2016
6
Special Report – Economics: Money supply – The forgotten indicator
Figure 3: China DM4 and real GDP
Figure 4: Euro-area Divisia M3 and real GDP
% y/y
% y/y
30
16
China DM4
14
25
12
10
12
20
10
15
8
10
6
China real
GDP (RHS)
2
0
0
2007
2009
2011
2013
2015
2
6
0
4
-2
Euro-area
DM3
2
-6
2006
2008
2010
Source: ChinaBond, Standard Chartered Research
Source: Bruegel, Eurostat, Standard Chartered Research
Figure 5: India Divisia M4 and real GDP
Figure 6: Japan DM3 and real GDP
% y/y
% y/y
2012
4.5
India Divisia
M4
20
2014
DM3
4.0
25
4
2
0
Japan real
GDP (RHS)
1.5
1.0
India CPI
India real GDP
0
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
-8
-0.5
-10
-1.0
-12
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Source: Bloomberg, Economic and Social Research Institute Japan
Figure 7: UK Divisia M4 and real GDP
Figure 8: US Divisia M4 and real GDP
% y/y
% y/y
12
6
10
4
UK real GDP
(RHS)
8
15
10
0
5
4
-2
0
2
-4
0
-6
UK DM4
-2
2003
2006
2009
Source: Bloomberg, BoE, Standard Chartered Research
2012
US DM4 incl
USTs
2
6
10 March 2016
-4
0.0
Source: India Central Statistical Organisation, RBI, Standard Chartered Research
2000
-2
-6
0.5
5
6
3.0
2.0
10
8
3.5
2.5
15
-4
0
2004
30
4
8
4
5
6
Euro-area
Real GDP
(RHS)
-8
2015
-5
US real GDP
-10
1968 1971 1975 1979 1983 1986 1990 1994 1998 2001 2005 2009 2013
Source: Bloomberg, Centre for Financial Stability, BEA
7
Special Report – Economics: Money supply – The forgotten indicator
Why money supply is important
Theory and history
For transactions, people need
money
For more than 200 years economic theory has recognised a close link between
money and GDP. One way of thinking of this is that for people to make transactions
they need to get the money together first. Another is that people try to keep the ratio
of their money holdings to other assets broadly constant, so if they accumulate more
money relative to other assets they will seek to spend it to restore equilibrium. That
said, economists since Keynes have often disputed the direction of causation
between money and GDP. And since the failure of the monetarist experiments in the
1980s, many have doubted the practicality of using money as an indicator,
particularly since financial innovation continues to generate new instruments that can
be used for transactions (see Appendix 1 for more on the historical context).
The quantity theory of money
The clearest formulation of the quantity theory is usually known as the Fisher equation
MV = PT
or
Money * Velocity = Prices * Transactions
Velocity (V) is the number of times money changes hands. Note that transactions (T)
include not just goods and services, i.e., GDP. It includes all transactions including
financial transactions. Hence a rise in money may portend an increase in stock or
house purchases. This effect can be seen in some of our analysis – e.g., the jump in
US money growth in 1987 ahead of the stock-market crash of that year and again
during the stock and housing bubbles around the turn of the century.
While strictly speaking the Fisher equation is an accounting identity, not a causal
relationship, classical theory (and monetarists) insists that a rise in M will flow
through to more transactions (and eventually higher prices). Some analysts argue
that an expanding money supply might have little effect because velocity may simply
decline to offset it. However, this view goes against more than two centuries of theory
and experience. That said, the links between money supply and the economy come
with what economist Milton Friedman called ‘long and variable lags’.
Simple sum money measures are unstable
The practical problem with using money supply as an indicator is that since the
1980s SSMs have proven relatively unstable and not well-correlated with GDP
trends. China is one of the few countries that still uses a money supply target for
reference; most central banks pay little attention to money supply. Modern financial
systems have too many instruments that have some degree of moneyness. Simple
sum aggregation can provide misleading signals.
Divisia may be the answer
Divisia money could overcome
some of the shortfalls of SSMs
The use of Divisia measures of money addresses these problems and provides more
confidence we are looking at a meaningful measure of money. This may be
especially important amid current low and even negative interest rates, as the
usefulness of interest rates as an indicator of the economy or policy stance is in
question. For example, few people now are confident that low rates mean that
monetary policy is super-stimulatory and that growth will surge. We need an
alternative way to look at monetary conditions.
10 March 2016
8
Special Report – Economics: Money supply – The forgotten indicator
Money supply and quantitative easing
QE needs to be enough to take
money growth to a healthy rate
Another reason for looking at money supply is that central banks are now employing
QE as a stimulus technique. Many of them view QE primarily as a way to signal their
intention to keep interest rates low for a long time by bringing down long-term interest
rates. But its success as well as the inflation risk involved can also be judged by what
happens to money growth. If broad money growth picks up to a healthy rate the amount
of QE is probably about right. What is a healthy rate? For a developed economy
somewhere in the 4-7% range is probably about right, in our view. For China and India
a range of 10-15% may be enough to support GDP growth of 6-8%, moderate inflation
and some financial deepening.
QE does not necessarily boost money supply
There is a more or less 1-1 relationship between central bank purchases and M0 or
so-called central bank money. But broader measures of money from M1-M4 only
include money in circulation, which is what affects GDP. And reserves tucked away
at the central bank (included only in M0) are not in circulation. This is why the big rise
in M0 since 2009 in the US and other developed countries has not brought either an
economic boom or higher inflation.
Unfortunately, when central banks purchase securities from banks there is no direct
impact on the money supply beyond M0. Banks simply substitute reserves at the
central bank for the securities they have sold. QE only raises money in circulation if
the securities are purchased from non-bank institutions such as pension funds or
insurance companies. The central bank directly credits those institutions’ bank
accounts with newly created money.
QE may have been inadequate
QE was not enough to compensate
for deleveraging
We believe our Divisia estimates give us a good way of assessing whether the right
amount of QE has been done. Evidence suggests that if anything, quantitative easing
has been less than required since 2009. Money growth in the US, UK and euro area
remains far less buoyant than pre-GFC. This is likely partly because too much was
purchased from banks. Also QE needed to be enough to compensate for
deleveraging. Especially in the immediate aftermath of the GFC, the private sector
was paying down debt and banks were raising capital and trying to reduce their
balance sheets. These actions have the effect of reducing the money supply.
The good news is that money growth has picked up in the past year, notably in the
euro area where QE began in Q1-2015. But the euro area and Japan will likely need
to continue QE for a long time yet, and we cannot rule out further QE from the US
and UK if growth disappoints. Money growth in the US is at the low end of our
‘healthy’ rate of 4-7%.
Figure 9: Money supply components as a share of GDP
% of nominal GDP, 2015
M0
M1
M2
M3
M4
China
8.7
56.0
198.4
215.4
219.8
India
10.6
17.4
17.8
80.4
81.7
Euro area
10.3
99.7
136.6
148.0
NA
Japan
18.2
126.4
237.2
245.1
NA
United Kingdom
4.0
-
-
-
84.7
United States
7.6
17.1
69.0
-
-
Source: Bloomberg, IMF, Standard Chartered Research
10 March 2016
9
Special Report – Economics: Money supply – The forgotten indicator
How Divisia money is calculated
Weighting money components by their usefulness as money
Divisia money assesses the degree
of ‘moneyness’ of different
components
Conventional measures of money supply are calculated by simply adding together
the components, ranging from currency and demand deposits in M1, time deposits
and money market funds in M2 and instruments such as CDs, commercial paper
and short-term debt securities for broader measures such as M3 and M4 (see
Appendix 2). The moneyness of each asset is assumed to be the same.
In reality the degree of moneyness varies, depending on ease and convenience of
use and reliability as a store of value. Currency and checking accounts are the
purest form of money and are mostly held for transactions. Using time deposits for
payments may incur a penalty or a delay or involve some inconvenience (such as
visiting a branch). Often time deposits are intended to be held for the longer term.
Securities such as commercial paper or short-term bonds can usually be sold
easily but may require time for settlement and their value could change due to
market or credit conditions.
A theoretically robust method
A theoretically robust weighting
methodology
Divisia money is a theoretically robust method for weighting the components of
money supply according to their degree of moneyness. The technique is named
after French economist Francois Divisia (1889-1964) and has been developed and
championed primarily by US economist William Barnett. The weights are calculated
according to the interest rate paid on each component; those which are less useful
for transactions will enjoy a higher rate. Cash and checking accounts which pay
zero interest have the highest weight while relatively illiquid assets that earn closer
to the benchmark rate have the lowest weight. The weights will change as the
structure of interest rates change, allowing Divisia money to compensate for shifts
among components.
How to construct a Divisia money index
To calculate a Divisia index for money requires the following:
Simple sum measures have been
found to be unstable

The amounts outstanding, and interest rates earned for each component.

A ‘benchmark’ interest rate against which the rate for each component can be
compared. This benchmark asset should enjoy the typical interest rate return in
the economy but be relatively illiquid and not be used for transactions.
The infographic on page 4 includes a simple illustration of how to calculate M2,
assumed to consist of cash, checking accounts and time deposits, where the first two
earn zero interest, time deposits earn 2% and the benchmark rate is 4%. M2 simply
sums the three components. Divisia M2 includes only half the time deposits
(calculated by taking 4/2). In practice the calculation is made using growth rates, so
the weighting of the growth of time deposits is cut in half.
If interest rates fell to 1% on time deposits and 3% on the benchmark asset, a higher
weight is given to the growth of time deposits, two-thirds. So is the money supply
suddenly larger? It depends, because people will respond to different interest rates
by moving assets from one component to another. When interest rates fall people
hold relatively more in demand deposits, because there is less advantage to keeping
money in time deposits.
10 March 2016
10
Special Report – Economics: Money supply – The forgotten indicator
A comparison of trends in US M1 and M2 with the federal funds target rate shows
this behaviour clearly: M1 and M2 move in opposite directions when interest rates
shift, more in M1 when rates go down and more in M2 when rates go up (Figure 10).
This shift in money supply measures is one of the key reasons why simple sum
money was found to be unstable in the past. The other was new instruments, which
could be used for money.
Choosing the benchmark rate
The benchmark rate must be higher
than the rates on all the
components
In practice it is hard to find an asset that exactly matches the requirements for the
benchmark asset; most assets have some degree of liquidity, while those that have
less, such as small stocks or private equity, involve high risk and volatility and are
hard to value. Moreover, since the benchmark rate must not be below the rate paid
on the money components (as negative moneyness makes no sense), this rules out
using a long-term bond yield as the yield curve can sometimes be inverted.
One approach is to use a bank lending rate (as we do for India). Another approach
formerly used by the BoE is to arbitrarily add a spread to a rate, in this case local
government bills plus 200bps. Yet another technique, which the BoE now uses,
assumes that the benchmark asset is the component of money supply that pays
the highest interest rate at any given time.
Despite these difficulties, research suggests that the choice of benchmark rate is less
important than it may appear because it is primarily used to measure the relative
moneyness of different components. The trend of Divisia growth rates turns out to be
not much affected by the choice of benchmark rates (Barnett 2011).
Figure 10: M1 and M2 go in opposite directions when rates change
US M1, M2 less M1 % y/y, and the federal funds target rate %
25
20
15
M1
10
5
FFTR
0
M2 - M1
-5
-10
1990
1994
1998
2002
2006
2010
2014
Source: US Federal Reserve, Standard Chartered Research
10 March 2016
11
Special Report – Economics: Money supply – The forgotten indicator
Why Divisia money is superior
Better in theory
In summary, Divisia monetary aggregates are better in principle than simple sum
counterparts for four reasons:
1.
The simple sum approach implies that all components of money stock are
substitutes or alike in their moneyness, which is plainly not true.
2.
Simple sum fails to take into account substitution effects in the event of a change
in the relative interest rates on different monetary assets (as discussed in the
example above). Simple sum aggregates especially do not cope well with
extreme situations such as inverted yield curves or QE policies.
3.
Using Divisia instead of simple sum overcomes many of the distortions of new
monetary instruments because the degree of moneyness of the new instruments
is automatically taken into account.
4.
Using relative interest rates to derive the weights is a theoretically consistent
approach.
Better in practice too
Divisia money performs better than
simple sum aggregates for most
countries
Studies covering at least 10 countries have found that Divisia money aggregates are
better than simple sum (Figure 11). Divisia money is not a magic crystal ball but it is
generally found to be more reliably correlated with GDP than simple sum aggregates.
Moreover, several studies find that it adds useful information for forecasters and nowcasters (gauging the state of the economy ahead of other indicators). Our
investigations also find that Divisia money performs better than simple sum
aggregates for most countries, though the stories vary.
Our approach
In our country analysis, we focus on the linkages between changes in money growth
and changes in real GDP. This is a simplification of the theory that suggests that the
linkage should be between nominal money growth and nominal domestic demand
(GDP plus imports less exports). We present this data in Figures 28-33 for those
interested. But gauging nominal trends in demand does not tell us the breakdown
between real demand and prices. We believe it is better to view money growth as first
influencing real growth, then after time it may influence inflation. For example weaker
money growth is likely to first slow real demand growth, then lower the inflation rate.
What about the distinction between domestic demand and GDP? This matters
periodically, when a country sees a big change in its net export picture. For example, it
is likely to be important for the US this year because the strong USD means that
aggregate demand will likely be higher than GDP. But the difference is usually fairly
small. Some analysts prefer to compare real money growth (after subtracting inflation)
with real GDP (see Figures 34-39).
10 March 2016
12
Special Report – Economics: Money supply – The forgotten indicator
Figure 11: Past studies suggest Divisia is better than SSMs
Studies producing Divisia
Monetary Aggregates
Summary
1980
Economic Monetary Aggregates: An
Application of Aggregation and Index
Number Theory – William Barnett
Barnett’s pioneering derivation of the Divisia money formula and
computation and discussion of the use of Divisia monetary
aggregates.
1984
The New Divisia Monetary Aggregates – First empirical comparisons of Divisia monetary aggregates with
William Barnett, Edward Offenbacher
SSM in policy applications; Divisia money, especially at higher levels
and Paul Spindt
of aggregation, usually performs better than simple sum when tested
against conventional policy-relevant criteria.
2011
How Better Monetary Statistics Could
Have Signaled the Financial Crisis –
William Barnett and Marcelle Chauvet
Most recessions in the past 50 years were preceded by
contractionary monetary policy, which is better indicated by Divisia
than by SSM data. Divisia monetary aggregate growth rates were
generally lower than SSM growth rates in the period preceding the
mid-1980s and higher since.
2005
Divisia Money – Matthew Hancock
BoE produces an improved measure of UK Divisia money and argues
for its usefulness.
2014
Does Money Matter in the Euro Area?
Evidence from a New Divisia Index –
Zsolt Darvas (Bruegel)
Money matters for output, prices and interest rates and the ECB can
influence monetary developments. VAR models reveal significant
euro-area output and price responses to Divisia money shocks.
Divisia money reacts to user costs. Most of these results are not
significant when SSM is used.
2015
Chinese Divisia Monetary Index and
GDP Nowcasting – William Barnett and
Biyan Tang
Divisia monetary aggregates contain more indicator information than
SSM aggregates; factor model produces the best available nowcasting results. Divisia money showed that China’s money supply
declined at the beginning of 2010 after which the growth rates of
Divisia M1, M2, M3 and M4 all steadily decreased, reflecting the
tightened monetary conditions in China.
2010
Empirical evidence suggests that Divisia monetary aggregates might
The Divisia Monetary Indices as Leading have an edge over SSM in a liberalised financial market regime in
Indicators of Inflation – M.
predicting inflation. As countries such as India continue to liberalise
Ramachandran, R. Das and Binod Bhoi their interest rate structure and deepen their financial systems,
Divisia measures will become more important.
2009
Divisia monetary aggregates are superior over SSM in that they
Comparison of Simple Sum and Divisia perform better for the financial innovation period of the early 1980s
Monetary Aggregates Using Panel Data and 1990s. Divisia seems to overcome simple sum aggregates’
Analysis – S. Celik and S. Uzun
inability to react to financial innovation and provide a stable money
demand function.
Indonesia
2010
Financial Liberalisation and Money
Demand in Indonesia: Implications for
Weighted Monetary Aggregates – Hiew
Lee Chea
Looks at Indonesia’s monetary regime changes and the significance
of Divisia monetary aggregates in shaping monetary policy in
Indonesia from 1981-2005. Empirical findings indicate that Divisia
monetary aggregates have proven to be superior to SSM as a useful
measure of money for Indonesia.
Malaysia
2008
Revisiting Money Demand in Malaysia:
Simple-Sum Versus Divisia Monetary
Aggregates – Chin-Hong Puah, ChoiMeng Leong et.al
Attempts to determine whether Divisia monetary aggregates are
superior monetary instruments in Malaysia where monetary targeting
has been replaced by interest rate targeting due to the inadequacy of
SSMs in predicting economic activities.
Germany
2015
The Information Content of Monetary
Statistics for the Great Recession:
Evidence from Germany – Wenjuan
Chen and Dieter Nautz
Introduces a Divisia monetary aggregate for Germany. Divisia money
and SSM are highly correlated in normal times but began to diverge
before the 2008 crisis, suggesting that Divisia was a better predictor
of the crisis.
UAE
2012
Divisia Monetary Aggregates for the
GCC Countries – Ryadh Alkhareif and
William Barnett
Divisia indexes are superior to the officially published SSM
aggregates in monitoring business cycles. Divisia money provides
critical information to policy makers regarding liquidity conditions.
2009
An Empirical Study of Simple Sum and
Divisia Monetary Aggregation: A
Comparison of their Predictive Power
Regarding Prices and Output in Turkey –
Dogan Karaman (PhD thesis, University
of Kansas)
Uses several models and methods to compare SSM and Divisia
aggregates in predicting Turkish inflation and output growth in the
last two decades; overall, money provides a good amount of
information in predicting inflation and output in Turkey. In highinflation periods, Divisia aggregates clearly provide better information
than SSM, while in low-inflation periods simple sum aggregates are a
better predictor.
Country
Year
United States
United Kingdom
Euro area
China
India
Japan
Turkey
Source: Centre for Financial Stability, Standard Chartered Research
10 March 2016
13
Special Report – Economics: Money supply – The forgotten indicator
Country studies
Global – Picking up but still lacklustre
Our aggregate is heavily influenced by China
Our global index shows weak
money growth during 2009-11
We aggregate the six major countries in this study to create a global index and
compare it with a global index using simple sum money in each country and also with
GDP trends (all using the same aggregation methodology, see below). We find that
Divisia matches GDP trends better than SSMs especially during the global financial
crisis. In 2008-09, when Divisia money growth declined sharply in line with the
economic slowdown, simple-sum money actually accelerated, reflecting quantitative
easing (Figure 12).
From 2009-11 Divisia money growth was weak, in line with the weak global recovery.
It briefly approached a 7% growth rate in 2012 before slowing again. The most recent
reading shows Divisia money growth of 7.6% y/y, the fastest since 2008 (Figure 13).
That said, a comparison with 2008 needs to allow for the hugely increased weight for
China: 21% in 2015 versus 8% in 2007. With China’s Divisia money growing at 15%
compared with developed country rates of 4-6%, the aggregate rate of growth is
hiding big changes in composition. (Calculated at China’s 2006 weight, global Divisia
money would be growing at only 6.2%). This is another reminder of the increase in
importance of policy and performance in China compared with less than 10 years
ago. Still, we view the acceleration in our global Divisia aggregate since 2014 as
positive, as it reflects a significant pick-up in the euro area and China.
A note on the aggregation methodology
We smooth exchange rate moves to
avoid confusion
The six countries here cover 68% of world GDP and all the major reserve currencies.
To weight countries, we use market exchange rates but smooth changes over three
years to avoid sharp exchange rate moves dominating money supply moves.
Otherwise a sudden change in the euro (EUR) could significantly affect the global
money supply in a way that we doubt makes sense in analysing the relationship
between money growth and GDP. This method still allows us to capture broad shifts
in economic gravity. Since 2007 the euro-area weight declined from 31% to 25%, the
US from 39% to 35% and China increased from 8% to 21%. The rise of China
reflects mainly its GDP growth and to a lesser extent exchange rate appreciation.
The EUR exchange rate changes little over the period using our smoothing method.
The large decline in weight reflects its poor GDP performance.
Figure 12: Our global Divisia estimate
Figure 13: Global Divisia and GDP growth
Divisia and simple sum money
% y/y
10
12
Divisia
8
10
6
8
SSM
4
Global GDP
growth index
2
6
DM
0
4
-2
2
0
2007
-4
2008
2009
2010
Source: Standard Chartered Research
10 March 2016
2011
2012
2013
2014
2015
-6
2007
2008
2009
2010
2011
2012
2013
2014
2015
Source: Standard Chartered Research
14
Special Report – Economics: Money supply – The forgotten indicator
China – Fastest rate in three years
We publish new broad money statistics
We present estimates for simple
sum and Divisia M3 and M4 not
published by the authorities
Official money supply statistics from the Peoples Bank of China (PBoC) provide data
for M1 and M2. We construct new broad money measures for China covering M3 and
M4 based on a research paper by Barnett (2015). The official M2 aggregate includes
currency in circulation and corporate, personal and other deposits. Our wider
definitions take into account debt instruments up to three years, including commercial
bank bonds, corporate bonds, commercial paper, local and central government
bonds and also money market funds. The broader definitions are comparable to the
broader definitions available elsewhere, including in the UK, US, and euro area.
Divisia is a better fit with GDP than official M2
For China, official M2 and our calculations of Divisia M4 mostly track fairly closely but
there is one major exception, in 2012 (Figures 14 and 15). Official M2 bounced
strongly in that year, yet GDP did not, suggesting that official M2 gave an inferior
signal. Barnett’s analysis also found that Divisia aggregates contain more indicator
information than simple sum aggregates and were better at now-casting China GDP.
Divisia M4 differs comparatively little from simple sum M4 (both calculated by us). But
China has liberalised interest rates only recently and plans further financial-sector
liberalisation. We expect this to lead to more flexibility and eventually bring the
potential for greater divergence between Divisia and SSMs. That said DM4 was
stronger than simple M4 in 2009-10, which probably more accurately reflected the
strength of the economy then. In general as financial markets deepen in emerging
markets, we expect Divisia money to diverge more from SSMs. For this reason it will
be increasingly important to monitor Divisia.
Divisia money is picking up
China’s Divisia money growth has
picked up since mid-2015,
suggesting the economy may
stabilise
Money supply growth, both the official M2 and our Divisia M4 estimate, has picked up
since the middle of 2015, suggesting that the economy may stabilise this year. This
supports our view that a hard landing is unlikely and our above-consensus GDP
forecast of 6.8%. The government has cut interest rates and reserve requirements,
as well as stoked fiscal policy and we expect more stimulus in coming months. China
has also relaxed macro-prudential policies, bringing a rise in housing-market activity
(see Special Report, 23 November 2015, ‘Asian housing: Ride the cycle’).
Figure 14: China Divisia M4 vs official M2
Figure 15: China Divisia M4 vs real GDP
% y/y
% y/y
30
30
25
20
China DM4
15
China M2
2009
2011
Source: ChinaBond, Standard Chartered Research
10 March 2016
25
25
20
20
China M4
15
15
10
10
5
2007
30
China DM4
2013
2015
5
0
2007
10
5
China real
GDP
0
2009
2011
2013
2015
Source: ChinaBond, Standard Chartered Research
15
Special Report – Economics: Money supply – The forgotten indicator
Euro area – A strong bounce in 2015
Divisia money grew very slowly during 2010-14
Euro area Divisia grew slowly
during 2010-14, underlining weak
monetary conditions
Divisia euro-area money supply at times differs quite markedly from its SSM
counterpart, in terms of both profile and growth rate (Figure 16). In the 2004-07
period Divisia M3 did not accelerate as much as official M3, making it a better fit with
real GDP growth, which was only mildly up (Figure 17). The European Central Bank
(ECB) raised rates in 2006-07 and even hiked again in July 2008, despite the
worsening financial crisis. The main fear was that higher oil prices would feed into
inflation expectations, but the double-digit growth in simple M3 may have been a
concern too. The Divisia measure, in contrast, peaked at 10% in September 2007
and slowed significantly by mid-2008.
Both measures did decline significantly during and after the GFC. This is different to
the experience in the UK and US, where simple sum surged initially and could be
because the ECB did not do quantitative easing at the time. During 2010-14 Divisia
money grew slowly, especially compared with pre-GFC growth, underlining the
weakness of monetary conditions. This sluggishness suggests that the ECB should
have begun QE long before it did in January 2015.
Divisia is now growing at 7%
Both measures of M3 have accelerated in the past 18 months, with Divisia money
now growing about 7%. This puts money growth only slightly below the 2003-07
average and likely reflects both active QE and the recent rise in lending to the private
sector. But GDP growth has been lacklustre and most indicators suggest it will
remain modest: Our 2016 GDP forecast of 1.4% is close to consensus. In contrast
strong Divisia money growth could be pointing to a growth surprise.
ECB studies (2001 and 2010) and others have concluded that Divisia monetary
aggregates display favourable econometric properties. The Bruegel Institute
publishes quarterly estimates of euro-area Divisia M3. Its analysis comparing Divisia
money with simple-sum measures reaches a number of conclusions: Divisia
aggregates have impacts on euro-area output and prices, output is impacted three to
nine quarters after a Divisia-money shock, the output shock is temporary, and the
ECB in practice reacts to developments in monetary aggregates. When measured via
SSM, the results are less meaningful.
Figure 16: Euro-area M3 vs Divisia M3
Figure 17: Euro-area M3 vs real GDP
% y/y
% y/y
14
14
12
12
10
Euro-area
DM3
8
10
8
Euro-area
DM3
6
6
4
4
2
Euro-area
M3
2
0
-2
2004
Euro-area M3
0
Euro-area
real GDP
-2
-4
2006
2008
2010
Source: Bruegel, Standard Chartered Research
10 March 2016
2012
2014
-6
2004
2006
2008
2010
2012
2014
Source: Bruegel, Eurostat, Standard Chartered Research
16
Special Report – Economics: Money supply – The forgotten indicator
India – Running slowly since 2012
Divisia M4 diverges significantly from simple M4
Divisia suggests an improvement in
India’s economic growth
Our analysis finds that Divisia M4 (DM4) shows a closer correlation with GDP than
does simple M4 (published by the Reserve Bank of India, ‘RBI’). The two series
diverge in four periods (Figure 18). From late 2006 to 2008 signs of overheating in
India’s economy are better captured by the Divisia measure, which was running
higher. During this period India sustained GDP growth at over 8.5% and the RBI
tightened policy to control inflation.
In 2008-09 the growth of Divisia money fell sharply, which seems to fit well with the
sharp economic slowdown at the time. In contrast, simple M4 remained high during
the crisis, giving misleading signals. It likely picked up the effects of the RBI’s policy
actions (multiple cuts in the cash reserve ratio and interest rates) to inject liquidity
and arrest India’s declining growth rate. In April 2009 the growth in DM4 started to
pick up, reflecting the economic recovery, before falling sharply when India’s growth
slowed again. In contrast, simple sum M4 only rose and fell slightly.
Divisia picks up trends in inflation too
India differs from the other countries in this study in that there was a major change in
the inflation rate during the period. Wholesale price inflation of 10-11% in 2010-12
came down to the 5-6% range in 2014 before falling even more in response to lower
commodity prices. Consumer price inflation, the central bank’s target now, has also
fallen significantly from 10-12% in 2009-13 to 5-6% now (Figure 19). We believe our
Divisia index is broadly consistent with this trend. Moreover the volatility in our
measure of DM4 seems to pick up the volatility of consumer prices during 2011-14,
falling sharply, then spiking then falling again.
Since 2012 both measures have moved broadly sideways in line with GDP, though
DM4 has been on a slight strengthening path while simple M4 has been slowing a bit.
Divisia is signalling an improvement in economic growth or at least suggests a
continuing moderate pace.
Figure 18: India M4 and real GDP growth
Figure 19: India Divisia M4 and headline inflation
% y/y
% y/y
30
30
25
25
20
20
15
M4
DM4
10
15
India DM4
10
India CPI
5
Real GDP
0
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Source: RBI, Bloomberg, CEIC, Standard Chartered Research
10 March 2016
5
0
2005
2007
2009
2011
2013
2015
Source: RBI, Bloomberg, Standard Chartered Research
17
Special Report – Economics: Money supply – The forgotten indicator
Japan – Up with Abenomics
Divisia differs little from simple money
We are not convinced that faster
growth in Japan’s Divisia points to
stronger GDP growth
Data limitations and the flatness of Japan’s yield curve mean that when we estimate
Divisia money differentiation between assets is limited. The series therefore differs
little from simple money (Figure 20). That said, the trends in Japanese money are
hard to interpret. Japan’s economy grew quite solidly during 2003-07 (averaging
about 1.75% p.a. despite the declining labour force) and deflation even eased up,
with core CPI climbing to zero in early 2008. Yet money growth slowed in this period
and even went negative in 2007 (Figure 21).
The expected relationship with GDP growth is missing
One possible reason for this weak money growth is that the Bank of Japan (BoJ)
ended its first QE programme in 2005 (it began in 2001) and cut the size of its
balance sheet in 2006). Still, the expected linkage to GDP is lacking for Japan. So,
while faster growth of Divisia money in the last two years is encouraging, it is
difficult to be confident that this points to stronger GDP growth or higher inflation.
Moreover, despite substantial QE, with the BoJ balance sheet much larger as
percentage of GDP than the that in the US or euro area, the fact that broad money
is still growing ‘only’ 4% suggests scope for an even faster pace of QE.
Figure 20: Japan DM3 vs M3
Figure 21: Japan DM3 and real GDP
% y/y
% y/y
5
4
8
6
DM3
3
SM3
2
1
5
DM3
4
4
4
2
3
3
0
-2
-4
2
Real GDP
(LHS)
-6
2
1
1
-8
0
-10
-1
-1
2004 2005 2006 2007 2008 2009 2009 2010 2011 2012 2013 2014 2015
-12
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
-1
Source: Bloomberg, Standard Chartered Research
Source: Bloomberg, Economic and Social Research Institute Japan
0
10 March 2016
18
Special Report – Economics: Money supply – The forgotten indicator
UK – Slower recently
In the UK Divisia money seems to fit much better with economic trends than SSMs
such as M4 (Figures 22 and 23). During the early 2000s upswing DM4 remained
solid in the 7-10% range from 2001 until late 2007. In contrast, simple M4 showed a
distinct uptrend in 2005-07. Yet economic growth did not accelerate in this period.
Headline inflation and wages picked up a little but house-price growth slowed. It is
not clear that the acceleration in simple M4 was a useful signal. The base rate was
raised significantly in 2006-07 but monetary policy may have been too lax throughout
the period, as is implied by DM4 growing at a high 7-10%, rather than that monetary
tightening was especially warranted in 2006-07.
Divisia has performed especially well since 2007
Since 2007, DM4 has been a better
indicator in the UK
Since 2007 DM4 has been a much better indicator than simple M4. DM4 growth
began to slow in 2007 and slumped to near zero well ahead of the Lehman
bankruptcy, correctly signalling the recession. In contrast, simple M4 slowed slightly
in 2007 but remained high and then soared during 2008-09, giving misleading signals
as it picked up the effects of liquidity provision and quantitative easing.
In 2010 DM4 recovered to growth of about 5%, reflecting the initial economic
upswing, before falling back again when UK growth slowed during the 2011-12 euro
crisis. Simple M4 slumped to below zero in early 2011, again distorted by financial
conditions. Since then it has roughly followed the same profile as DM4 but whereas
DM4 has averaged growth around 6%, consistent with the solid recovery since then,
simple M4 has ranged around 0%. The most recent Divisia growth rate is slightly
lower than two years ago, again consistent with the slowdown to around 2% GDP
growth currently from 3% previously.
Figure 22: UK Divisia M4 vs real GDP
Figure 23: UK Divisia M4 vs real GDP
% y/y
% y/y
12
20
UK DM4
10
15
8
6
2
5
0
0
-2
UK M4
-4
-5
UK real GDP
-6
-8
1991
UK Divisia M4
10
4
1994
1997
2000
2003
2006
Source: Bloomberg, BoE, Standard Chartered Research
10 March 2016
2009
2012
2015
-10
2000
2005
2010
2015
Source: Bloomberg, BoE, Standard Chartered Research
19
Special Report – Economics: Money supply – The forgotten indicator
US – Stuck in low gear
Money supply is largely ignored
The strength in US DM4 suggests
economic growth will continue
in 2016
In the US simple money aggregates like M2 have proven unhelpful in understanding
the economy especially in the short and medium term, and neither the Federal
Reserve (Fed) nor most private economists pay much attention to these measures.
The Fed no longer even publishes broader measures of money, such as M3.
Yet Divisia M4 (DM4) calculated by the Centre for Financial Stability does seem to
perform much better than M2 historically (see Barnett 2011). In particular, the two
measures diverged sharply during two crucial episodes: the monetarist experiment in
1979-82 and the GFC (Figures 24-25).
DM4 correctly signalled the double-dip recessions in 1980-82
The US monetarist experiment beginning in November 1979 aimed to bring inflation
down gradually without a recession by targeting reduced money growth. It was
abandoned in November 1982 after two recessions in two years, which occurred
despite buoyant M2 growth. But DM4 did signal both recessions, suggesting that by
focusing on simple money measures, the Fed erred with overly tight policy.
Simple M2 growth gave false signals in 2008-09 and 2012
The most recent big divergence has been since 2007 (Figure 26). At the height of the
crisis in late 2008, M2 soared, reflecting the various emergency liquidity measures
taken by the Fed and the first round of QE beginning in November 2008. But DM4
collapsed, reflecting the economic slump, a much more useful signal. M2 surged to
show growth of almost 10% in the first half of 2012, suggesting a robust recovery and
even supporting some inflation concerns. DM4 was much weaker, particularly in H1
2012, correctly signalling the sluggish nature of the recovery and the need for QE3,
which duly began in September 2012.
Divisia money is still growing slower than pre-2008
In the past two years DM4 has picked up a bit but at 4% y/y is still at the floor of our
suggested healthy range of 4-7%. It is also well below the 6-8% range prevailing pre2008. The recent pick-up suggests a new recession, as some have feared, will likely
be avoided. But the strength of the US dollar means that some spending in the US
will leak abroad. Overall the weakness of Divisia money growth combined with the
strong USD supports our below-consensus GDP forecast of 1% for 2016.
Figure 24: US official M2 vs Divisia M4
Figure 25: US official M2 vs Divisia M4 since Jan 2006
% y/y
% y/y
15
12
US M2
10
10
US M2
8
6
4
5
2
0
0
-5
-2
US DM4 exUSTs SA
-4
-6
-8
US DM4 exUSTs SA
-10
1968 1973 1978 1983 1988 1993 1998 2003 2008 2013
-10
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Source: Bloomberg
Source: Bloomberg
10 March 2016
20
Special Report – Economics: Money supply – The forgotten indicator
Divisia and the USD
We find a good relationship between money trends and the major multi-year swings
in the US dollar over the past 40 years (Figure 27). When money growth is
accelerating, the trade-weighted index tends to be rising and vice-versa. The intuition
here is straightforward. Accelerating money supply growth occurs when the economy
is strengthening; this also tends to be when the US dollar is strong.
However the relationship between money growth and exchange rates is complex
because it depends on whether the change in money growth is caused by
increased money demand (as in the US case) or increased money supply (as in
the case of Japan in recent years). Japan’s QE programme, promising faster
money growth, led to a sharp fall in the exchange rate. Because of these
complications and also because a full analysis of exchange rates requires
looking at relative money growth between different countries, we leave the
subject for future discussion.
Figure 26: US Divisia M4 vs real GDP growth
Figure 27: US Divisia M4 vs USD trade-weighted index
% y/y
15
140
US DM4 excl
Treasuries
10
US real TWI
120
15
10
100
5
0
-5
US real GDP
-10
1968 1973 1978 1983 1988 1993 1998 2003 2008 2013
Source: Bloomberg, Centre for Financial Stability, BEA
10 March 2016
80
5
60
0
40
20
US DM4 incl
USTs 2 yr mov
avg (RHS)
0
1973 1976 1980 1984 1987 1991 1995 1998 2002 2006 2009 2013
-5
-10
Source: Bloomberg, Centre for Financial Stability, BEA
21
Special Report – Economics: Money supply – The forgotten indicator
Why don’t central banks use Divisia?
Despite the evidence for its usefulness, only the UK’s central bank regularly
publishes Divisia data. While others have looked at it, Divisia has yet to catch on in a
major way either in central banks or academic circles. Money supply is out of fashion.
We think there are six reasons for this:
1.
Modern economic theory, the so-called Keynesian-Neoclassical synthesis, uses
only a short-term interest rate to link the financial sector to the real economy,
(though this proved hopelessly inadequate in 2008 when models based on this
theory had no way to incorporate a financial crisis).
2.
When theorists go beyond the interest rate they tend to focus on credit more
than money. Former Fed Chairman Bernanke based his prior academic career
on the case that credit rather than money is the key to the economy. And the
Bank for International Settlements, which is critical of standard theory, also
emphasises the credit cycle rather than money trends.
3.
The monetarist experiments in the late 1970s and early 1980s left an institutional
distaste in many central banks for focusing on money. They believe that the
inflation-targeting regime, perhaps supplemented by macro-prudential
measures, is the best approach. And, with fears that money trends can be
misleading, especially in the short and medium run, central banks are reluctant
to draw much attention to money.
4.
Concerns remain that financial innovation can confuse trends and that Divisia
money might not neutralise the effect.
5.
Calculations of Divisia money can be hard to explain and may appear to be a
‘black box’. See Appendix 2 for a further discussion of this issue
6. Even if Divisia money has outperformed in the past, it remains to be seen
whether it can still work well with low or negative interest rates and flat yield
curves. Our research suggests that it is effective, but it is early days. The case of
Japan is a concern.
Conclusion – Add Divisia to your dashboard
Central banks should pay more
attention to Divisia money
Our findings suggest that using Divisia estimates instead of simple-sum measures
significantly improves money growth as an indicator. The method of estimating
makes it a theoretically more trustworthy indicator of trends in money growth and it
can be used as a guide both to monetary policy stance and GDP trend.
Significant acceleration or deceleration in Divisia growth rates (2-3ppt or more) seem
to presage moves up or down in the GDP growth rate. Particularly slow money
growth, of 4% or below, seems to correlate with disappointingly slow GDP growth,
while fast growth rates of 7% or more in developed countries (as prior to 2008) or
rates of 20% or more in China and India are associated with overheating economies.
Divisia is not a magic crystal ball, but it looks like a useful indicator to monitor
alongside the usual closely watched activity and survey data. We hope central banks
will pay more attention as more data on the components of money supply and the
interest rates they earn could improve the precision of Divisia estimates. We plan to
provide regular updates and keep a close eye on Divisia.
10 March 2016
22
Special Report – Economics: Money supply – The forgotten indicator
Figure 28: China Divisia and domestic demand
Figure 29: Euro area Divisia and domestic demand
% y/y, annual avg Divisia growth rate
% y/y
30
15
25
10
20
5
DM4
15
Euro area
DM3
0
-5
10
Domestic
demand
5
0
2008
-10
Domestic
demand
-15
2009
2010
2011
2012
2013
2014
2004
2006
2008
2010
2012
2014
Source: ChinaBond, National Bureau of Statistics, Standard Chartered Research
Source: Eurostat, Standard Chartered Research
Figure 30: India Divisia and domestic demand
Figure 31: Japan DM3 and nominal aggregate demand
% y/y, INR bn
% y/y
18
2,500
16
6
4
2,000
14
12
1,500
India DM4
10
5
Domestic
demand
(RHS)
2
DM3
3
0
SM3
-2
8
1,000
6
Domestic
demand
(RHS)
4
2
500
0
2012
2013
2014
-4
1
0
-10
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
2015
Source: RBI, Bloomberg, Standard Chartered Research
Source: Ministry of Finance Japan, Standard Chartered Research
Figure 32: UK Divisia and domestic demand
Figure 33: US DM4 and domestic demand
% y/y
% y/y
12
20
UK DM4
2
-6
-8
0
2011
4
-1
15
15
10
10
8
5
6
US DM4 incl
Treasuries
10
5
0
4
2
0
-2
2000
-5
Domestic
demand
(RHS)
-10
-5
Domestic
demand
-15
-20
2003
2006
Source: Eurostat, Standard Chartered Research
10 March 2016
0
2009
2012
2015
-10
2000 2001 2002 2003 2005 2006 2007 2008 2010 2011 2012 2013 2015
Source: Bloomberg, Standard Chartered Research
23
Special Report – Economics: Money supply – The forgotten indicator
Figure 34: China real Divisia and real GDP
Figure 35: Euro-area real Divisia and real GDP
% y/y
% y/y
30
10
8
25
6
20
4
Real Divisia
15
Real GDP
2
0
10
-2
5
0
2007
Real Divisia
Real GDP
2008
2009
2010
2011
2012
2013
2014
2015
-4
-6
-8
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Source: ChinaBond, Standard Chartered Research
Source: Bloomberg, Standard Chartered Research
Figure 36: India real Divisia and real GDP
Figure 37: Japan real Divisia and real GDP
% y/y
% y/y
25
8
6
20
Real Divisia
(using WPI)
15
Real Divisia
4
2
0
10
-2
Real GDP
5
-6
-8
0
-5
2005
Real GDP
-4
-10
2007
2009
2011
2013
2015
-12
2004
2006
2008
2010
2012
Source: India Central Statistical Organisation, RBI, Standard Chartered Research
Source: Bloomberg, Economic and Social Research Institute Japan
Figure 38: UK real Divisia and real GDP
Figure 39: US real Divisia and real GDP
% y/y
% y/y
10
2014
15
8
Real Divisia
10
6
4
Real Divisia
5
2
Real GDP
0
0
-2
-4
Real GDP
-5
-6
-8
2000
2003
2006
2009
Source: Bloomberg, BoE, Standard Chartered Research
10 March 2016
2012
2015
-10
1968 1973 1978 1983 1988 1993 1998 2003 2008 2013
Source: Bloomberg, Centre for Financial Stability, BEA
24
Special Report – Economics: Money supply – The forgotten indicator
Appendix 1: The quantity theory of money
Quantity theory used to be central to economics
Central bankers lost faith in the
simple measures of money supply
Until the 1930s the quantity of money was a major focus of economics. Famous
monetary theorists including Hume, Ricardo, Marshall, Fisher and Keynes (in his
earlier writings) argued that a rise in the money supply led to increased economic
growth in the short run, though in the long run GDP would fall back leaving only a rise
in the price level. The workings of the Gold Standard were much studied and seen as
operating through changes in the money supply and in wages.
The publication of Keynes’ ‘General Theory’ in 1936 transformed economic theory by
suggesting that, at least in times of inadequate aggregate demand, wages might not
adjust properly and even active monetary policy (freed from the constraints of the
Gold Standard) might be incapable of restoring growth (Keynes 1936). Instead
expansionary fiscal policy would be needed. Theory after Keynes has continued to
emphasise the key role of aggregate demand. Changes in interest rates, which would
influence aggregate demand, became both the instrument and measure of monetary
policy, in place of the quantity of money.
The monetarist revival
For a brief period in the 1970s and early 1980s the quantity theory came back to the
forefront of economics during the monetarist experiments. Central banks in the US,
UK and Europe tried to target the growth of money supply with the aim of gradually
bringing down high inflation without a recession. For a few years money supply
releases became the most-watched economic indicator, moving markets in a way
that employment or purchasing managers indices do today.
Inflation did come down (under the guidance of Paul Volcker at the Fed and Margaret
Thatcher in the UK) but only at the expense of severe recessions and high
unemployment. Meanwhile the targeted quantities (principally M2 in the US and
Sterling M3 in the UK), proved unpredictable and unstable. Within a few years
monetarism was abandoned and mainstream theory shifted back to the focus on
interest rates, increasingly ignoring money supply trends.
Why did monetarism fail?
There were several reasons for the failure of monetarism. If a particular measure of
money is targeted it becomes unreliable as an indicator of economic trends (known as
Goodheart’s law after monetary theorist Charles Goodheart). This is because investors
know it is targeted and seek to make money from the expected central bank response,
which distorts the outcome. Meanwhile, during the monetarist experiments interest
rates were high and volatile, again changing money demand and supply in ways that
models were unable to predict. Finally the 1970s and 1980s was also the beginning of
a period of rapid financial innovation, partly due to deregulation but also in response to
high interest rates, high inflation and monetary targeting itself. New liquid assets
appeared and holdings grew rapidly, including money market funds and various shortterm securities. Central bankers lost faith in the meaning of simple measures of money
supply and concentrated instead on interest rates.
In the 2010s interest rates are no longer a reliable signal
In past cycles, low interest rates were a clear signal that monetary policy was
expansionary and the economy should strengthen. Today, amid fears that demand is
so weak that interest rates cannot be pushed low enough, low interest rates are not
necessarily a signal of coming expansion, or a reliable policy instrument.
10 March 2016
25
Special Report – Economics: Money supply – The forgotten indicator
QE and the money supply
QE has played an important role but
more may be needed
Although central banks have actively used quantitative easing, they have generally
understood this in terms of interest rates not the quantity of money, despite the
name. (QE is only the modern name for what used to be called ‘open market
operations’, a standard policy in the past.) In the US QE1 was widely seen as an
exercise to bring down credit spreads, particularly on mortgage securities, as a way
of boosting growth. It is generally seen as a major success.
QE2 and QE3, introduced when credit spreads were more normal, have been heavily
criticized and remain controversial, as does ongoing QE in Japan and the euro area.
Standard theory takes it that the driver of spending decisions is expected future real
short-term interest rates. So QE is seen as a tool to help drive down expected future
interest rates and is used alongside inflation targets and explicit forward guidance, as
the BoJ and ECB do today.
Central banks have therefore looked at the impact of QE on long-term interest rates
as an indicator of success, since long-term rates can be seen as the sum of future
short term rates. But if QE is expected to boost growth prospects and future inflation
then it should raise long-term bond yields, not cut them. So trying to gauge the
success of QE by looking for a fall in bond yields is flawed.
Using money supply to judge QE effectiveness
Since bond yields are misleading, a better approach theoretically is to look at the
impact on money supply. Here some analysts made a basic mistake. They saw the
rise in the monetary base, which is caused by QE (on essentially a one-for-one
basis), as hugely inflationary. But the monetary base consists largely of reserves
tucked away at the central bank. These excess reserves have no influence on the
economy (and are not included in money supply measures such as M1 to M4).
Classical theory is quite clear that these broader measures, that is money in
circulation, are what matters.
QE only directly increases the money supply (M1 upwards) in two circumstances.
One is if the bonds are purchased from non-bank entities such as pension funds or
insurance companies, in which case the central bank pays the seller in newly created
money. The other is if the bonds are purchased from banks and the banks respond
with new lending and/or purchases of securities from non-banks. Unfortunately, if
central banks purchase bonds from banks who simply replace them with reserves at
the central bank, there is no increase in broad money and no economic effect.
In principle, a good way to measure whether QE is effective is to see whether broad
money is expanding. Bear in mind that if the private sector is deleveraging or banks
are ultra-conservative, broad money could decline. As people repay bank loans or
banks raise their capital ratios banks shrink both their assets and liabilities. Hence
QE has to be enough to offset this if money supply is to rise. As discussed above,
our findings are that money supply growth, properly measured by Divisia, has picked
up but remains well below pre-2007 levels. This suggests that QE has played an
important role but more may be needed.
10 March 2016
26
Special Report – Economics: Money supply – The forgotten indicator
Appendix 2: Calculating Divisia money
Calculating Divisia
Divisia money is calculated as a weighted average of the growth rate of the different
components of money supply. The weights for the components are determined by
their usefulness for making transactions, which is proxied by the user cost of holding
these assets.
The user cost of a component is measured by the difference between the benchmark
rate, which is the post-tax interest rate paid on balances with no use for transactions,
and the post-tax interest rate paid on the component balance. Divisia money supply
growth is therefore calculated as follows:
∆Dt
N
=
Dt–1 ∑
i=1
1
∆Mi,t
2 (Wi,t + Wi,t–1) Mi,t–1
Wi,t =
Mi,t (rB,t – ri,t)
N
∑ Mi,t (rB,t – ri,t）
i=1
Where:
Mi = the level of the ith money holding
Wi = the weight on the ith component
rB = the rate on the benchmark asset
ri = the rate on the ith asset.
Figure 40: Definitions of money supply
Lowest measure of money supply that includes component
Component
Currency in circulation
Demand deposit/other
checkable deposits
Savings deposits
China
Euro area
India
Japan
UK
US
M0
M0
M0
M0
M0
M0
M1
M1
M1
M1
M4
M1
M1
M1
M4
M3
M1
M4
M2
Time deposits
M2
Foreign currency deposits
M3
M2 (deposits
redeemable at a
period of notice up
to 3 months)
Other deposits
Money market funds
Repurchase agreement
M3 (SC)
M2 (quasi-money)
M4
M4
M2 (money
market accounts,
retail money
market funds), M3
(institutional
money market
mutual funds)
M3
M4
M2
M3
M3
M2 (post office
savings deposits),
M4 (other post
office deposits)
Postal savings deposit
Certificate of deposit
Debt instruments
M3
M3
M3 (SC): Central
bank bonds, Policy
bonds, Commercial
bank bonds,
M3 (debt
corporate bonds, securities up
commercial paper, to 2 years)
T-bills), M4 (SC):
Treasury bonds,
local government
M4 (bonds,
commercial paper,
FRNs, instruments
M4-(commercial
of up to and
paper), M4
including five
(T-bills)
years original
maturity, bank
bills)
Source: Standard Chartered Research
10 March 2016
27
Special Report – Economics: Money supply – The forgotten indicator
Divisia money growth produced by Standard Chartered Bank (SC)
Figure 41: China
Money supply
Components
Interest rate for Divisia calculation
Source
M0 (official)
Currency in circulation
Assumed zero
People’s Bank of China
M0
Corporate demand deposits
M1
Corporate deposits
Personal deposits
Other deposits
M2
Central bank bonds
Policy bank bonds
Commercial bank bonds
Corporate bonds
Commercial paper
Money market funds
M3
Treasury bonds
Local government bonds
Corporate demand
deposit
People’s Bank of China
3-month time deposit
People’s Bank of China
Central bank bill rate
Inter-bank policy bank bond yield
Inter-bank commercial bank bond yield
Chinabond corporate bond yield
Inter-bank commercial paper yield
One-day interbank repo-rate
Inter-bank treasury bond yield
Chinabond local government
security yield
China Central Depository and
Clearing Corporation Limited
M1 (official)
M2 (official)
M3 (SC)
M4 (SC)
Benchmark rate
WIND
Bloomberg
China Central Depository and Clearing
Corporation Limited
Short term corporate bond yield + 100 bps
Source: Standard Chartered Research
Figure 42: India
Money supply
Components
Interest rate for Divisia calculation
Source
M0 (official)
Currency in circulation
Assumed zero
Reserve Bank of India (RBI)
Prime lending rate
RBI
M1 (official)
M2 (official)
M0
Deposit money of the public
(includes demand deposits
with Banks and other deposits
with Banks)
M1 + Post Office Savings
Deposits
Post Office savings
Bank account rate
1 year time deposit rate
Average of 1yr and 3 yr time deposit
rates
3 yr time deposit rate
5 yrs and above time deposit rate
M3 (official)
M1 + time deposits with banks
M4 (official)
M3 + total Post Office deposits Post Office time deposit account rate
Benchmark rate
Prime lending rate
RBI
State Bank of India
RBI
Source: Standard Chartered Research
Figure 43: Euro area
Money supply
M1 (official)
M2 (official)
M3 (official)
Benchmark rate
Components
Interest rate for Divisia calculation
Source
Currency in circulation
Overnight deposits
M1
Deposits with agreed maturity
of 2 years
Deposits redeemable at a
period of notice up to 3 months
M2
Repurchase agreements
Money market funds
Debt securities up to 2 years
Currency rate assumed zero
Overnight deposit rate
European Central Bank (ECB)
The rate on deposits with agreed
maturity of 2 years
The rate on deposits redeemable at
notice of up to three months
ECB
The rate on repurchase agreements
Eonia rate
BofA Merrill Lynch 1-3 year Euro
Financial Index
ECB
Bloomberg
BofA Merrill Lynch 1-3 year Euro Financial Index + 100 bps
Source: Standard Chartered Research
10 March 2016
28
Special Report – Economics: Money supply – The forgotten indicator
Figure 44: Japan
Money supply
Components
M0
Currency in circulation
M1
M3
Currency in circulation + demand deposits
M1 + quasi-money (household)
Quasi-money (corporate)
M2 + CDs
Benchmark rate
Average rate on short-term loans, city banks
M2
Interest rate for Divisia
calculation
Assumed zero
Assumed zero
2-3 yr deposit rate
3m deposit rate
CD rate
Source
Bank of Japan
Source: Standard Chartered Research
Divisia money growth produced by others
Figure 45: UK
Money supply
M0 (official)
M4 (official)
Components
Currency in circulation
Banks operational deposits with the Bank of England
M0
Deposits (including certificate of deposits)
Commercial paper
Bonds
FRNs and other instruments of up to and including five years
of original maturity issued by UK MFIs
Repos
Bank bills
Divisia
Divisia M4 produced by Bank of England
Source: Standard Chartered Research
Figure 46: US
Money supply
Components
M0 (official)
Currency in circulation
M1 (official)
M2 (official)
M3 (Centre for
Financial Stability)
M4- (Centre for
Financial Stability)
M4 (Centre for
Financial Stability)
M0
Demand deposits
Travellers’ checks
Checkable deposits
M1
Money market accounts
Retail money market funds
Certificate of deposits
M2
Large time deposits
Institutional money market mutual funds balances
Deposits of euro dollars
Repurchase agreements
M3
Commercial paper
M4T-bills
Divisia
Centre for Financial Stability produces Divisia M1,
Divisia M2, Divisia M3 and Divisia M4
Source: Standard Chartered Research
10 March 2016
29
Special Report – Economics: Money supply – The forgotten indicator
Practical problems in calculating Divisia money
While we have made every effort to calculate useful data there are some practical
difficulties in estimating Divisia, where more central bank data could help:
1.
Data availability issues. Full breakdowns of different instruments and their
interest rates are not always available, especially for long time series.
Sometimes simplifying assumptions have to be made, for example applying the
interest rate on one-year time deposits to all time deposits.
2.
Bundled services. The interest rate may not always be the full return on a
particular instrument. For example checking accounts often provide free
transaction services and automatic overdraft facilities. Time deposits sometimes
have complex introductory teaser rates or even gifts (from toasters to telephone
minutes).
3.
Choice of the benchmark interest rate. The Benchmark interest rate is often
difficult to choose or not easy to fully justify. That said, Divisia growth money
trends (if not rates) are not unduly sensitive to this because what matters is the
relative moneyness of each asset.
4.
No allowance for adjustment time and costs. As discussed above, when
interest rates change, Divisia money automatically changes the weights on
components but in practice, asset holdings are only adjusted over time,
which may be partly due to inertia and partly to costs of switching. Divisia
assumes that assets holdings are always at their desired values. This means
that trends in Divisia can only be watched over time; short-term moves may
be misleading.
Figure 47: Divisia and our forecasts
% y/y
Divisia money growth Divisia money growth
(latest)
2012-14 average
GDP forecast 2016
Consensus forecast
2016
China
15.4
14.2
6.8
6.5
Euro area
7.7
3.5
1.4
1.6
India
11.8
10.3
7.6
7.4
Japan
3.5
2.9
1.0
0.9
United Kingdom
5.4
5.2
1.9
2.1
United States
3.8
3.3
1.0
2.2
Source: Standard Chartered Research
10 March 2016
30
Special Report – Economics: Money supply – The forgotten indicator
References
Barnett W and Chauvet M, Financial Aggregation and Index Number Theory, World Scientific, 2011.
Barnett W and Biyan Tang, ‘Chinese Divisia Monetary Index and GDP Nowcasting’, University of Kansas and
Centre for Financial Stability, New York, October 2015.
Darvas Z, ‘Does Money Matter in the Euro Area? Evidence from a New Divisia Index’, Bruegel Working Paper
2014/12, November 2014 (updated April 2015).
Hancock M, ‘Divisia money’, Bank of England Quarterly Bulletin, 2005.
Ramachandran M and Rajib Das et. al., ‘The Divisia Monetary Indices as Leading Indicators of Inflation’,
Department of Economic Analysis and Policy, Reserve Bank of India, 2010.
10 March 2016
31
Special Report – Economics: Money supply – The forgotten indicator
Global Research Team
Management Team
Dave Murray, CFA +65 6645 6358
Marios Maratheftis +971 4508 3311
Head, Global Research
[email protected]
Standard Chartered Bank, Singapore Branch
Chief Economist
[email protected]
Standard Chartered Bank
Thematic Research
Madhur Jha +44 20 7885 6530
Enam Ahmed +44 0207 885 7735
Samantha Amerasinghe +44 20 7885 6625
Senior Economist, Thematic Research
[email protected]
Standard Chartered Bank
Senior Economist, Thematic Research
[email protected]
Standard Chartered Bank
Economist, Thematic Research
[email protected]
Standard Chartered Bank
Global Macro Strategy
Eric Robertsen +65 6596 8950
Mayank Mishra +65 6596 7466
Head, Global Macro Strategy and FX Research
[email protected]
Standard Chartered Bank, Singapore Branch
Macro Strategist
[email protected]
Standard Chartered Bank, Singapore Branch
Economic Research
Africa
Asia
Razia Khan +44 20 7885 6914
David Mann +65 6596 8649
Chief Economist, Africa
[email protected]
Standard Chartered Bank
Chief Economist, Asia
[email protected]
Standard Chartered Bank, Singapore Branch
Victor Lopes +44 20 7885 2110
Southeast Asia
Senior Economist, Africa
[email protected]
Standard Chartered Bank
Greater China
Shuang Ding +852 3983 8549
Head, Greater China Economic Research
[email protected]
Standard Chartered Bank (HK) Limited
Edward Lee Wee Kok +65 6596 8252
Kelvin Lau +852 3983 8565
Head, ASEAN Economic Research
[email protected]
Standard Chartered Bank, Singapore Branch
Senior Economist, HK
[email protected]
Standard Chartered Bank (HK) Limited
Jeff Ng +65 6596 8075
Betty Rui Wang +852 3983 8564
Economist, SEA
[email protected]
Standard Chartered Bank, Singapore Branch
Economist, NEA
[email protected]
Standard Chartered Bank (HK) Limited
Chidu Narayanan +65 6596 7004
Se Yan +86 10 5918 8302
Economist, Asia
[email protected]
Standard Chartered Bank, Singapore Branch
Senior Economist, China
[email protected]
Standard Chartered Bank (China) Limited
Usara Wilaipich +662 724 8878
Lan Shen +86 10 5918 8261
Senior Economist, Thailand
[email protected]
Standard Chartered Bank (Thai) Public Company Limited
Economist, China
[email protected]
Standard Chartered Bank (China) Limited
Aldian Taloputra +62 21 2555 0596
Tony Phoo +886 2 6603 2640
Senior Economist, Indonesia
[email protected]
Standard Chartered Bank, Indonesia Branch
Senior Economist, NEA
[email protected]
Standard Chartered Bank (Taiwan) Limited
South Asia
Korea
Anubhuti Sahay +91 22 6115 8840
Chong Hoon Park +82 2 3702 5011
Head, South Asia Economic Research
[email protected]
Standard Chartered Bank, India
Head, Korea Economic Research
[email protected]
Standard Chartered Bank Korea Limited
Senior Economist, Latam
[email protected]
Standard Chartered Bank NY Branch
Saurav Anand +91 22 6115 8845
Kathleen B. Oh +82 2 3702 5072
Economist, South Asia
[email protected]
Standard Chartered Bank, India
Economist, Korea
[email protected]
Standard Chartered Bank Korea Limited
Europe
Middle East and North Africa
Sarah Hewin +44 20 7885 6251
Dima Jardaneh +971 4 508 3591
Bilal Khan +92 21 3245 7839
Chief Economist, Europe
[email protected]
Standard Chartered Bank
Head of Economic Research, MENA
[email protected]
Standard Chartered Bank
Senior Economist, MENAP
[email protected]
Standard Chartered Bank (Pakistan) Limited
Achilleas Chrysostomou +44 20 7885 6437
Carla Slim +971 4 508 3738
Philippe Dauba-Pantanacce +44 20 7885 7277
Economist, Europe
[email protected]
Standard Chartered Bank
Economist, MENA
[email protected]
Standard Chartered Bank
Senior Economist, Global Political Analyst
[email protected]
Standard Chartered Bank
Sarah Baynton-Glen +44 20 7885 2330
Economist, Africa
[email protected]
Standard Chartered Bank
Edward Cheng +44 20 7885 5284
Economist, Africa
[email protected]
Standard Chartered Bank
Emmanuel Kwapong +44 20 7885 5840
Economist, Africa
[email protected]
Standard Chartered Bank
The Americas
Mike Moran +1 212 667 0294
Head, Economic Research, The Americas
[email protected]
Standard Chartered Bank NY Branch
Thomas Costerg +1 212 667 0468
Senior Economist, US
[email protected]
Standard Chartered Bank NY Branch
Italo Lombardi +1 212 667 0564
10 March 2016
32
Special Report – Economics: Money supply – The forgotten indicator
FICC Research
Rates Research
Credit Research
FX Research
Kaushik Rudra +65 6596 8260
Kaushik Rudra +65 6596 8260
Eric Robertsen +65 6596 8950
Head, Rates & Credit Research
[email protected]
Standard Chartered Bank, Singapore Branch
Head, Rates & Credit Research
[email protected]
Standard Chartered Bank, Singapore Branch
Head, Global Macro Strategy and FX Research
[email protected]
Standard Chartered Bank, Singapore Branch
Nagaraj Kulkarni +65 6596 6738
Shankar Narayanaswamy +65 6596 8249
Robert Minikin +44 20 7885 8674
Senior Asia Rates Strategist
[email protected]
Standard Chartered Bank, Singapore Branch
Head, Credit Strategy & Financials
[email protected]
Standard Chartered Bank, Singapore Branch
Head, Asian FX Strategy
[email protected]
Standard Chartered Bank
Arup Ghosh +65 6596 4620
Bharat Shettigar +65 6596 8251
Eimear Daly +44 20 7885 6162
Senior Rates Strategist
[email protected]
Standard Chartered Bank, Singapore Branch
Head, IG/Sovereign Credit Research
[email protected]
Standard Chartered Bank, Singapore Branch
G10 FX Strategist
[email protected]
Standard Chartered Bank
Lawrence Lai +65 6596 8261
Jaiparan Khurana +65 6596 7251
Nick Verdi +1 646 845 1279
Asia Rates Strategist
[email protected]
Standard Chartered Bank, Singapore Branch
Credit Analyst, IG/Sovereign
[email protected]
Standard Chartered Bank, Singapore Branch
Senior FX Strategist
[email protected]
Standard Chartered Bank NY Branch
Hee Eun Lee +65 6596 8690
Simrin Sandhu +65 6596 6281
Devesh Divya +65 6596 8608
Asia Rates Strategist
[email protected]
Standard Chartered Bank, Singapore Branch
Senior Credit Analyst, Financials & ME Corporates
[email protected]
Standard Chartered Bank, Singapore Branch
Asia FX Strategist
[email protected]
Standard Chartered Bank, Singapore Branch
Becky Liu +852 3983 8563
Nikolai Jenkins, CFA +65 6596 8259
Eddie Cheung +852 3983 8566
Senior Asia Rates Strategist
[email protected]
Standard Chartered Bank (HK) Limited
Credit Analyst, Financials
[email protected]
Standard Chartered Bank, Singapore Branch
Asia FX Strategist
[email protected]
Standard Chartered Bank (HK) Limited
John Davies +44 20 7885 7640
Zhi Wei Feng +65 6596 8248
Lemon Zhang +65 659 69498
US Rates Strategist
[email protected]
Standard Chartered Bank
Head, HY Credit Research
[email protected]
Standard Chartered Bank, Singapore Branch
[email protected]
Analyst, FX Research / Global Macro Strategy
Standard Chartered Bank, Singapore Branch
Samir Gadio +44 20 7885 8618
Chun Keong Tan, CFA +65 6596 8257
Head, Africa Strategy
[email protected]
Standard Chartered Bank
Credit Analyst, HY
[email protected]
Standard Chartered Bank, Singapore Branch
Eva Murigu +25 42 0329 4004
Jiacheng Chen +65 6596 8710
Africa Strategist
[email protected]
Standard Chartered Investment Services Kenya Limited
Credit Analyst, HY
[email protected]
Standard Chartered Bank, Singapore Branch
Flows Research
Melinda Kohar +65 6596 9543
Michael Trounce +44 20 7885 2058
Senior Strategist, Flows Research
[email protected]
Standard Chartered Bank
Credit Strategist
[email protected]
Standard Chartered Bank, Singapore Branch
Commodities Research
Paul Horsnell +44 20 7885 6913
Head, Commodities Research
[email protected]
Standard Chartered Bank
Nicholas Snowdon +44 20 7885 2276
Metals Analyst
[email protected]
Standard Chartered Bank
Yuhan Xia +44 207 8858670
Commodities Analyst
[email protected]
Standard Chartered Bank
Suki Cooper +1 212 667 0319
[email protected]
Precious Metals Analyst
Standard Chartered Bank NY Branch
Priya Balchandani +65 6596 8254
Energy Analyst
[email protected]
Standard Chartered Bank, Singapore Branch
Serene Shang Yi Lim +65 6596 6064
Energy Analyst
[email protected]
Standard Chartered Bank, Singapore Branch
Judy Zhu +86 21 6168 5016
Metals Analyst
[email protected]
Standard Chartered Bank, Singapore Branch
10 March 2016
33
Special Report – Economics: Money supply – The forgotten indicator
This page is intentionally blank.
10 March 2016
34
Special Report – Economics: Money supply – The forgotten indicator
Disclosures appendix
Analyst Certification Disclosure: The research analyst or analysts responsible for the content of this research report certify that: (1) the views expressed and attributed to the research analyst
or analysts in the research report accurately reflect their personal opinion(s) about the subject securities and issuers and/or other subject matter as appropriate; and, (2) no part of his or her
compensation was, is or will be directly or indirectly related to the specific recommendations or views contained in this research report. On a general basis, the efficacy of recommendations is a
factor in the performance appraisals of analysts.
Global Disclaimer: Standard Chartered Bank and/or its affiliates (“SCB”) makes no representation or warranty of any kind, express, implied or statutory regarding this document or any information
contained or referred to in the document. The information in this document is provided for information purposes only. It does not constitute any offer, recommendation or solicitation to any person to
enter into any transaction or adopt any hedging, trading or investment strategy, nor does it constitute any prediction of likely future movements in rates or prices, or represent that any such future
movements will not exceed those shown in any illustration. The stated price of the securities mentioned herein, if any, is as of the date indicated and is not any representation that any transaction
can be effected at this price. While reasonable care has been taken in preparing this document, no responsibility or liability is accepted for errors of fact or for any opinion expressed herein. The
contents of this document may not be suitable for all investors as it has not been prepared with regard to the specific investment objectives or financial situation of any particular person. Any
investments discussed may not be suitable for all investors. Users of this document should seek professional advice regarding the appropriateness of investing in any securities, financial
instruments or investment strategies referred to in this document and should understand that statements regarding future prospects may not be realised. Opinions, forecasts, assumptions,
estimates, derived valuations, projections and price target(s), if any, contained in this document are as of the date indicated and are subject to change at any time without prior notice. Our
recommendations are under constant review. The value and income of any of the securities or financial instruments mentioned in this document can fall as well as rise and an investor may get
back less than invested. Future returns are not guaranteed, and a loss of original capital may be incurred. Foreign-currency denominated securities and financial instruments are subject to
fluctuation in exchange rates that could have a positive or adverse effect on the value, price or income of such securities and financial instruments. Past performance is not indicative of comparable
future results and no representation or warranty is made regarding future performance. While we endeavour to update on a reasonable basis the information and opinions contained herein, there
may be regulatory, compliance or other reasons that prevent us from doing so. Accordingly, information may be available to us which is not reflected in this material, and we may have acted upon
or used the information prior to or immediately following its publication. SCB is not a legal, regulatory, business, investment, financial and accounting and/or tax adviser, and is not purporting to
provide legal, regulatory, investment, financial and accounting and/or tax advice. Independent legal, regulatory, business, investment, financial and accounting and/or tax advice should be sought
for any queries relating to the legal, regulatory, business, investment, financial and accounting and/or tax implications of any investment. SCB and/or its affiliates may have a position in any of the
securities, instruments or currencies mentioned in this document. SCB and/or its affiliates or its respective officers, directors, employee benefit programmes or employees, including persons
involved in the preparation or issuance of this document may at any time, to the extent permitted by applicable law and/or regulation, be long or short any securities or financial instruments referred
to in this document and on the SCB Research website or have a material interest in any such securities or related investments, or may be the only market maker in relation to such investments, or
provide, or have provided advice, investment banking or other services, to issuers of such investments and may have received compensation for these services. SCB has in place policies and
procedures and physical information walls between its Research Department and differing public and private business functions to help ensure confidential information, including ‘inside’ information
is not disclosed unless in line with its policies and procedures and the rules of its regulators. Data, opinions and other information appearing herein may have been obtained from public sources.
SCB expressly disclaims responsibility and makes no representation or warranty as to the accuracy or completeness of such information obtained from public sources. SCB also makes no
representation or warranty as to the accuracy nor accepts any responsibility for any information or data contained in any third party’s website. You are advised to make your own independent
judgment (with the advice of your professional advisers as necessary) with respect to any matter contained herein and not rely on this document as the basis for making any trading, hedging or
investment decision. SCB accepts no liability and will not be liable for any loss or damage arising directly or indirectly (including special, incidental, consequential, punitive or exemplary damages)
from the use of this document, howsoever arising, and including any loss, damage or expense arising from, but not limited to, any defect, error, imperfection, fault, mistake or inaccuracy with this
document, its contents or associated services, or due to any unavailability of the document or any part thereof or any contents or associated services. This material is for the use of intended
recipients only and, in any jurisdiction in which distribution to private/retail customers would require registration or licensing of the distributor which the distributor does not currently have, this
document is intended solely for distribution to professional and institutional investors.
Country-Specific Disclosures – If you are receiving this document in any of the countries listed below, please note the following:
United Kingdom and European Economic Area: SCB is authorised in the United Kingdom by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the
Prudential Regulation Authority. This communication is not directed at Retail Clients in the European Economic Area as defined by Directive 2004/39/EC. Nothing in this document constitutes a
personal recommendation or investment advice as defined by Directive 2004/39/EC. Australia: The Australian Financial Services Licence for Standard Chartered Bank is Licence No: 246833 with
the following Australian Registered Business Number (ARBN: 097571778). Australian investors should note that this communication was prepared for “wholesale clients” only and is not
directed at persons who are “retail clients” as those terms are defined in sections 761G and 761GA of the Corporations Act 2001 (Cth). Bangladesh: This research has not been produced in
Bangladesh. The report has been prepared by the research analyst(s) in an autonomous and independent way, including in relation to SCB. THE SECURITIES MENTIONED IN THIS REPORT
HAVE NOT BEEN AND WILL NOT BE REGISTERED IN BANGLADESH AND MAY NOT BE OFFERED OR SOLD IN BANGLADESH WITHOUT PRIOR APPROVAL OF THE REGULATORY
AUTHORITIES IN BANGLADESH. Any subsequent action(s) of the Recipient of these research reports in this area should be subject to compliance with all relevant law & regulations of
Bangladesh; specially the prevailing foreign exchange control regulations. Botswana: This document is being distributed in Botswana by, and is attributable to, Standard Chartered Bank
Botswana Limited which is a financial institution licensed under the Section 6 of the Banking Act CAP 46.04 and is listed in the Botswana Stock Exchange. Brazil: SCB disclosures pursuant to
the Securities Exchange Commission of Brazil (“CVM”) Instruction 483/10: This research has not been produced in Brazil. The report has been prepared by the research analyst(s) in an autonomous
and independent way, including in relation to SCB. THE SECURITIES MENTIONED IN THIS REPORT HAVE NOT BEEN AND WILL NOT BE REGISTERED PURSUANT TO THE REQUIREMENTS
OF THE SECURITIES AND EXCHANGE COMMISSION OF BRAZIL AND MAY NOT BE OFFERED OR SOLD IN BRAZIL EXCEPT PURSUANT TO AN APPLICABLE EXEMPTION FROM THE
REGISTRATION REQUIREMENTS AND IN COMPLIANCE WITH THE SECURITIES LAWS OF BRAZIL. China: This document is being distributed in China by, and is attributable to, Standard
Chartered Bank (China) Limited which is mainly regulated by China Banking Regulatory Commission (CBRC), State Administration of Foreign Exchange (SAFE), and People’s Bank of China (PBoC).
Germany: In Germany, this document is being distributed by Standard Chartered Bank Germany Branch which is also regulated by the Bundesanstalt für Finanzdienstleistungsaufsicht (BaFin). Hong
Kong: This document, except for any portion advising on or facilitating any decision on futures contracts trading, is being distributed in Hong Kong by, and is attributable to, Standard Chartered Bank
(Hong Kong) Limited 渣打銀行（香港）有限公司 which is regulated by the Hong Kong Monetary Authority. India: This document is being distributed in India by Standard Chartered Bank, India
Branch (“SCB India”). SCB India is a branch of SCB, UK and is licensed by the Reserve Bank of India to carry on banking business in India. SCB India is also registered with Securities and Exchange
Board of India in its capacity as Merchant Banker, Investment Advisor, Depository Participant, Bankers to an Issue, Custodian etc. For details on group companies operating in India, please
visit https://www.sc.com/in/india_result.html. The particulars contained in this document are for information purposes only. This document does not constitute an offer, recommendation or solicitation
to any person to execute any transaction with SCB India. Certain information or trade ideas in this document may not be specifically permissible under Indian regulations; hence, users of this
document should seek professional legal advice before acting on any information. Indonesia: The information in this document is provided for information purposes only. It does not constitute any
offer, recommendation or solicitation to any person to enter into any transaction or adopt any hedging, trading or investment strategy, nor does it constitute any prediction of likely future
movements in rates or prices or represent that any such future movements will not exceed those shown in any illustration. Japan: This document is being distributed to Specified Investors, as
defined by the Financial Instruments and Exchange Law of Japan (FIEL), for information only and not for the purpose of soliciting any Financial Instruments Transactions as defined by the FIEL or any
Specified Deposits, etc. as defined by the Banking Law of Japan. Kenya: Standard Chartered Bank Kenya Limited is regulated by the Central Bank of Kenya. This document is intended for use
only by Professional Clients and should not be relied upon by or be distributed to Retail Clients. Korea: This document is being distributed in Korea by, and is attributable to, Standard Chartered
Bank Korea Limited which is regulated by the Financial Supervisory Service and Financial Services Commission. Macau: This document is being distributed in Macau Special Administrative
Region of the Peoples' Republic of China, and is attributable to, Standard Chartered Bank (Macau Branch) which is regulated by Macau Monetary Authority. Malaysia: This document is being
distributed in Malaysia by Standard Chartered Bank Malaysia Berhad only to institutional investors or corporate customers. Recipients in Malaysia should contact Standard Chartered Bank Malaysia
Berhad in relation to any matters arising from, or in connection with, this document. Mauritius: Standard Chartered Bank (Mauritius) is regulated by both the Bank of Mauritius and the Financial
Services Commission in Mauritius. This document should not be construed as investment advice or solicitation to enter into securities transactions in Mauritius as per Securities Act 2005. New
Zealand: New Zealand Investors should note that this document was prepared for “wholesale clients” only within the meaning of section 5C of the Financial Advisers Act 2008. This document
is not directed at persons who are “retail clients” as defined in the Financial Advisers Act 2008. This document does not form part of any offer to the public in New Zealand. NOTE THAT SCB IS
NOT A “REGISTERED BANK” IN NEW ZEALAND UNDER THE RESERVE BANK OF NEW ZEALAND ACT 1989. Pakistan: The securities mentioned in this report have not been, and will not
be, registered in Pakistan, and may not be offered or sold in Pakistan, without prior approval of the regulatory authorities in Pakistan. Philippines: This document may be distributed in the
Philippines by, Standard Chartered Bank (Philippines) which is regulated by the Bangko Sentral ng Pilipinas (Telephone No. (+63) 708-7701, Website: www.bsp.gov.ph). This document is for
information purposes only and does not constitute, and should not be construed as an offer to sell or distribute in the Philippines securities that are not registered with the Securities and
Exchange Commission unless such securities are exempt under Section 9 of the Securities Regulation Code or such offer or sale qualifies as an exempt transaction under Section 10 thereof.
Singapore: This document is being distributed in Singapore by SCB Singapore branch and/or Standard Chartered Bank (Singapore) Limited, provided that research reports relating to certain products
may be distributed only to accredited investors, expert investors or institutional investors, as defined in the Securities and Futures Act, Chapter 289 of Singapore. Recipients in Singapore should
contact SCB Singapore branch or Standard Chartered Bank (Singapore) Limited (as the case may be) in relation to any matters arising from, or in connection with, this document. South Africa: SCB
is licensed as a Financial Services Provider in terms of Section 8 of the Financial Advisory and Intermediary Services Act 37 of 2002. SCB is a Registered Credit Provider in terms of the National
Credit Act 34 of 2005 under registration number NCRCP4. Thailand: This document is intended to circulate only general information and prepare exclusively for the benefit of Institutional
Investors with the conditions and as defined in the Notifications of the Office of the Securities and Exchange Commission relating to the exemption of investment advisory service, as amended
and supplemented from time to time. It is not intended to provide for the public. UAE: For residents of the UAE – Standard Chartered Bank UAE does not provide financial analysis or
consultation services in or into the UAE within the meaning of UAE Securities and Commodities Authority Decision No. 48/r of 2008 concerning financial consultation and financial analysis.
UAE (DIFC): SCB is regulated in the Dubai International Financial Centre by the Dubai Financial Services Authority. This document is intended for use only by Professional Clients and Market
Counterparties and should not be relied upon by or be distributed to Retail Clients. United States: Except for any documents relating to foreign exchange, FX or global FX, Rates or Commodities,
distribution of this document in the United States or to US persons is intended to be solely to major institutional investors as defined in Rule 15a-6(a)(2) under the US Securities Exchange Act of 1934.
All US persons that receive this document by their acceptance thereof represent and agree that they are a major institutional investor and understand the risks involved in executing transactions in
securities. Any US recipient of this document wanting additional information or to effect any transaction in any security or financial instrument mentioned herein, must do so by contacting a registered
representative of Standard Chartered Securities (North America) Inc., 1095 Avenue of the Americas, New York, N.Y. 10036, US, tel + 1 212 667 0700. WE DO NOT OFFER OR SELL SECURITIES
TO U.S. PERSONS UNLESS EITHER (A) THOSE SECURITIES ARE REGISTERED FOR SALE WITH THE U.S. SECURITIES AND EXCHANGE COMMISSION AND WITH ALL APPROPRIATE
U.S. STATE AUTHORITIES; OR (B) THE SECURITIES OR THE SPECIFIC TRANSACTION QUALIFY FOR AN EXEMPTION UNDER THE U.S. FEDERAL AND STATE SECURITIES LAWS NOR
DO WE OFFER OR SELL SECURITIES TO U.S. PERSONS UNLESS (i) WE, OUR AFFILIATED COMPANY AND THE APPROPRIATE PERSONNEL ARE PROPERLY REGISTERED OR
LICENSED TO CONDUCT BUSINESS; OR (ii) WE, OUR AFFILIATED COMPANY AND THE APPROPRIATE PERSONNEL QUALIFY FOR EXEMPTIONS UNDER APPLICABLE U.S. FEDERAL
AND STATE LAWS. Any documents relating to foreign exchange, FX or global FX, Rates or Commodities to US Persons, Guaranteed Affiliates, or Conduit Affiliates (as those terms are defined by
any Commodity Futures Trading Commission rule, interpretation, guidance, or other such publication) are intended to be distributed only to Eligible Contract Participants are defined in Section 1a(18)
of the Commodity Exchange Act. Zambia: Standard Chartered Bank Zambia Plc is licensed and registered as a commercial bank under the Banking and Financial Services Act Cap 387 of the laws of
Zambia and is regulated by the Bank of Zambia, the Lusaka Stock Exchange and the Securities Exchange Commission.
© Copyright 2016 Standard Chartered Bank and its affiliates. All rights reserved. All copyrights subsisting and arising out of all materials, text, articles and information contained herein is the
property of Standard Chartered Bank and/or its affiliates, and may not be reproduced, redistributed, amended, modified, adapted, transmitted in any form, or translated in a ny way without the
prior written permission of Standard Chartered Bank.
Document approved by
Document is released at
Marios Maratheftis
Chief Economist
14:12 GMT 10 March 2016
10 March 2016
35