Transcript Determinants of Financialisation in South Africa: A

Determinants of Financialisation in
South Africa: A Balance Sheet
Approach
Rex A McKenzie
October 3rd 2013
Pretoria, South Africa
Focus of the paper
• This paper focuses on the process of
finacialisation in South Africa. Financialisation is
defined after Krippner (2005) and Arrighi (1994),
as a pattern of accumulation in which profits
accrue through financial markets and channels
rather than through trade or commodity
production in the real economy. We examine the
financialisation of the South African economy
from 1990 to 2012—both before and after major
financial market liberalizations.
EDD/UNDP/UNDESA/CSID
2
Five key features
• First, a precipitous increase in the size of the
net acquisition of financial assets (and the net
incurrence of financial liabilities) on the
balance sheets of all agents.
EDD/UNDP/UNDESA/CSID
3
1. Net Acquisition of Financial Assets
800
700
600
R billion
500
Non-financial private sector
400
Households
Government
300
Foreign sector
200
100
0
199219931994199519961997199819992000200120022003200420052006200720082009201020112012
-100
EDD/UNDP/UNDESA/CSID
4
1. Net incurrence of financial liabilities
350000
300000
250000
200000
R billion
Net saving
150000
Consumption of fixed capital
Net incurrence of financial liabilities
Bank loans and advances
100000
Ordinary shares
50000
0
-50000
-100000
EDD/UNDP/UNDESA/CSID
5
2. JSE market capitalisation /portfolio
flows
• Third a measureable relationship between levels of Johannesburg
Stock Exchange (JSE) market capitalisation and levels of portfolio
capital inflows.
• Model 1: A 10% growth in market capitalisation induces a 26. 1%
increase in portfolio inflows. Further, there is a positive relationship
between growth in money supply and portfolio flows. The results
further shows that a 10% growth in money supply will result in an
approximately 5% increase in portfolio flows. Last, the results show
that there is a negative relationship between openness of the
economy and portfolio flows, a 10% increase in openness will
reduce portfolio flows by about 1.12%. The results tell us that
market size, interest rate and previous portfolio flows were all
insignificant determinants of portfolio flows in the period under
study.
EDD/UNDP/UNDESA/CSID
6
2. JSE market capitalisation /portfolio
flows
• Model 2: A 10% increase in market capitalisation
results in approximately 30% increase in foreign
portfolio investment. The model shows a positive
relationship between an increase in money
supply and portfolio flows; a 10% increase in
money supply results in a 4.4% increase in
portfolio flows. The model also shows that a 10%
depreciation of the rand results in 1.2% increase
in portfolio flows. Last, the results show that
there is a negative relationship between
openness of the economy and portfolio flows.
EDD/UNDP/UNDESA/CSID
7
3. Centrality of Credit and Foreign
Funds
• 2012, provides a good overview of the financial interconnections across the main sectors of the South
African economy in 2012. The largest recipient of
foreign sector funds in 2012 was the NFI sector,
followed by the general government sector. The
general government sector received R84 billion and the
NFI, R116 billion. The NFI’S subsequently extended R98
billion to the general government sector, which
allowed government to fund its net borrowing position
of R194 billion. The MFI’S overall held a net lending
position of R69 billion, from which it extended R66
billion to the NFI’S (Monyela and Nhelko, 2013).
EDD/UNDP/UNDESA/CSID
8
3. Centrality of Credit and Foreign
Funds
116 NFBS
(82)
197
Foreign Sector
66
2
98
MFI'S 2
69
General Govt.
1
10
(194)
84
1 Households
11
EDD/UNDP/UNDESA/CSID
9
4. Credit effects that overwhelm
portfolio shift effects
• Duc and Breton 2009; According to the authors;
“Variations in money holding by agents can be broken
down into changes in the total financial assets held by
agents and changes in the share of liquid assets within
this total amount.” The authors identify two main
sources of changes in the money holdings of non-MFI
sector; the first is a “credit effect” where the rise in
total financial assets of non-MFI is mainly due to credit.
The second is a “portfolio shift effect” representing the
shift between money and more illiquid financial assets
in the total financial assets held by non-MFI.
EDD/UNDP/UNDESA/CSID
10
4. Credit effects that overwhelm
portfolio shift effects
• The following identity formalises the
distinction:
• M ≡ (M/FA) × FA
• Where, M stands - money and
• FA - (total) financial assets.
• FA represents the “credit effect”, and M/FA
represents the “portfolio shift effect.”
EDD/UNDP/UNDESA/CSID
11
4. Credit effects that overwhelm
portfolio shift effects
90
80
70
60
50
portfolio shift effect
credit effect
40
30
20
10
0
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
EDD/UNDP/UNDESA/CSID
12
5. High degree of financial
intermediation
• Fifth, and last an unusually high degree of
financial intermediation (when compared with
the Euro zone area) that may be explained by
the high degree of concentration in South
Africa’s Monetary and Financial Institutions
Sector (MFI’s).
EDD/UNDP/UNDESA/CSID
13
High degree of financial
intermediation
• Aim is to assess developments in “market”
financing versus “bank financing” in the South
African economy. In other words, we want to
quantify and thereby measure intermediation
tendencies in the local economy.
EDD/UNDP/UNDESA/CSID
14
High degree of financial
intermediation
100
Intermediation ratio
90
80
70
60
50
40
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
EDD/UNDP/UNDESA/CSID
15
Conclusion
• From a policy point of view one big question emerges – given the
constellation of financial intermediation in South Africa, its unique
concentration, and the dominance of MFI credit effects in explaining
changes in the liquidity holdings of agents - what can we say about the
workings of monetary policy? Presumably the dominance of finance
houses, their credit products in particular, and their large shadow over the
macro-economy must have some impact on the workings of monetary
policy. A lesser question relates to the impact of debt on the balance sheet
of agents. Net incurrence implies debt, and as net incurrence grows so
does debt; at what level does the debt become excessive so as to become
a constraint on the agent and economic growth in the economy as a
whole? Definitive answers to these questions will not be forthcoming until
more studies and more research efforts are directed at these issues. For
now we have a basic roadmap that can be used to define a research
project focusing on the five key areas identified in the forgoing.
EDD/UNDP/UNDESA/CSID
16