Transcript Doing growth diagnostics in practice

Cost and access to capital as
constraints to growth
Elena Ianchovichina
PRMED, World Bank
Joint Vienna Institute
July 2009
Growth diagnostics
Problem: Low levels of private investment and entrepreneurship
High cost of finance
Low return to economic activity
Low appropriability
Low social returns
poor natural
resource
management
poor
geography
low
human
capital
government
failures
bad international
finance
market
failures
information
externalities:
“self discovery”
bad infra
structure
- micro risks:
macro risks:
property rights,
financial,
corruption,
monetary, fiscal
taxes
instability
bad local finance
-
coordination
externalities
low
domestic
saving
poor
inter
mediation
-
Source: Hausmann, Rodrik, Velasco (2005)
How do we assess whether a country is liquidity
constrained?


Misleading to rely on the popular measure of the
amount of credit to the private sector as a share of
GDP
Low quantity of credit to the private sector is not
necessarily a signal of scarcity of the factor


Quantity of finance may be low because of scarce supply,
in which case the country is considered liquidityconstrained (e.g. Brazil, 2000-07)
But it may be low because of low demand, in which case
the economy is not liquidity constrained, it is a case of
low returns (e.g. Zambia, 2000-07)
Domestic Credit to Private Sector (% of GDP) vs Per Capita
GDP. 2000-2007
Domestic Credit to private Sector / GDP
250
200
y = 20.291x - 110.49
R² = 0.5118
150
100
Mongolia
Benin
50
Brazil
Tajikistan
Zambia
0
4
5
6
7
8
9
LN(GDPpc Current US$)
10
11
12
What measures do we use?


The price can be used to distinguish between the two cases
In the case of finance, the price to look at is the real interest
rate


Low quantity and high price indicate scarcity of supply relative to
demand
Look at international comparisons and the distance of the price from
the mean


if the price signal is an outlier and is several standard deviations outside
the expected range it is difficult to reject the hypothesis
It is important to look at investment by sector – investment
may be adequate at the aggregate level but it may be
concentrated in one sector or a few sectors

Investment allocation tells us about the type of growth process
occurring in the country and the likelihood that growth will be broad
based and inclusive
(Domestic Credit to private Sector / GDP)*100
Domestic Credit to Private Sector (% of GDP) vs Real
Interest Rate. Constrained Countries 2000-2007
Mean Real Lending Rate = 8.4%
50
Mean Dom Cred to Priv / GDP = 49.5%
40
Mean Real Lending Rate + 1 std Dev = 15%
Brazil
30
Mongolia
20
Tajikistan
10
Benin
Zambia
(Mean Dom Cred to Priv / GDP) -1 Std Dev= 3.8%
0
0%
5%
10%
15%
20%
25%
30%
Real Lending Rates
35%
40%
45%
50%
Objective data

Look at objective data over time and cross-country:



Average real lending rates as a proxy of real cost of capital – time
series data
Benchmark to comparators
Bank’s lending rates by maturity and by borrower


Investigate the reasons for the change in the cost of capital




The range tells us a lot about the costs faced by different types of firms
Is it due to changes in inflation?
Is it due to changes in deposit rates?
Is it due to changes in risk premiums?
Benchmark deposit rates and risk premiums

What are the determinants of deposit rates and risk premiums?
Subjective data


Use firm survey data to see what are the perceptions about the
cost and access to capital
Match perceptions with reality, find out:




Distinguish between loans of different maturity


The percentage of firms complaining about the cost and access to
finance as a severe constraint to firms’ growth
The percentage of firms that did not apply for a loan because of the
high cost of capital
The percentage of firms that either obtained a loan or did not need a
loan
Access to long-term financing is typically a big problem
Even when the cost of capital is high, and the credit to the
private sector is low, if the majority of firms do not need
loans to expand operations then the country may not be
liquidity constrained
Why is a country liquidity constrained?

Inadequate access to savings

Both access to foreign borrowing and domestic savings
must be limited






High spreads on foreign borrowing due to high country risk and
low credit ratings
Domestic capital controls
Poverty traps
High tax burden
Important to understand who is not saving: firms or
households
Inefficient process of financial intermediation
Inefficient process of financial
intermediation

Difficulty assessing credit risk



May lead to very high risk premiums for the
majority of firms, especially small firms
High collateral requirements
Reasons



Poor corporate governance
Lack of transparency in business operations
Weakness in the bankruptcy and debt recovery
framework
Inefficient process of financial
intermediation

Issues with access to capital

Underdeveloped capital markets

The financial system may be dominated by a few banks, the stock
market may be illiquid, commercial bond market may not exist, and
pension funds may be small

Banks may offer a limited range of products

People may not be using the banking system due to lack of trust

Financial system misallocates resources



Could result in high opportunity costs and possibly high fiscal costs
Banks may lend for consumption but not for productive projects
Banks may lend to the elite and well-connected, and for this exclusive
group, private cost of capital may be low, while for the majority of
individuals access to capital may be very limited
Three case studies



Mongolia
Benin
Zambia
The case of Mongolia
Was Mongolia liquidity constrained in 2005?
Was private investment too low in Mongolia?

Gross domestic investment in Mongolia was
high for its level of development


Averaged 35% of GDP between 1996 and 2005
However, most of investment was official
foreign aid and loans

59% of investment in 2004
Was Mongolia liquidity constrained in 2005?
Did the composition of investment in Mongolia support inclusive
growth?

The bulk of private investment went into a limited
number of firms in mining and construction



FDI was high and averaged 5.2% of GDP in 1996-2005
Domestic private investment was financed mainly by own
funds (72 percent in 2004), and not bank loans
Domestic credit to the private sector was growing at high
rates, but most of the loans were short term and financed
trade, not productive investments
Were real interest rates in Mongolia high?
Real interest rates came down substantially…
80
70
60
%
50
40
30
20
10
0
1996
1997
1998
Mongolia
Source: Ianchovichina and Gooptu (2007)
1999
2000
Cambodia
2001
2002
Vietnam
2003
2004
2005
Kyrgyz Republic
What was the reason for the fall in the real cost of capital?
The fall in the real cost of capital was due to inflation rate increases rather than risk
premium declines…
18
16
14
12
10
8
6
4
2
0
-2
2001
2002
2003
2004
Interest rate spread (lending rate minus deposit rate)
Inflation, consumer prices (annual %)
Source: Ianchovichina and Gooptu (2007)
2005
and Mongolia’s cost of capital was still high relative to
other developing countries
Real interest rate
Mongolia
Source: Ricardo Hausman, “A framework for Growth Diagnostics”, Kennedy School of Government, Harvard University, May 2006.
Romania
Poland
Ecuador
Pakistan
Bosnia
Nicaragua
Bangladesh
Algeria
Moldova
India
Philippines
Slovakia
El Salvador
Czech
Kenya
Morocco04
Estonia
Bulgaria
Croatia
Morocco00
Indonesia
Russia
Egypt
Hungary
Albania
China
Brazil
Slovenia
Turkey
4
2
0
-2
-4
-6
Oman
14
12
10
8
6
and given the availability of credit
to the private sector
(Domestic Credit to private Sector / GDP)*100
Domestic Credit to Private Sector (% of GDP) vs Real
Interest Rate. Constrained Countries 2000-2007
Mean Real Lending Rate = 8.4%
50
Mean Dom Cred to Priv / GDP = 49.5%
40
Mean Real Lending Rate + 1 std Dev = 15%
Brazil
30
Mongolia
20
Tajikistan
10
Benin
Zambia
(Mean Dom Cred to Priv / GDP) -1 Std Dev= 3.8%
0
0%
5%
10%
15%
20%
25%
30%
Real Lending Rates
35%
40%
45%
50%
Why was the cost of capital still high in Mongolia?
Cost of capital was high because of high bank deposit rates and
risk premiums
25
20
15
10
5
0
Mongolia
Azerbaijan
Deposit interest rate
Cambodia
Kyrgyz
Republic
Vietnam
Uruguay
Interest rate spread (lending rate minus deposit rate)
Source: Ianchovichina and Gooptu (2007)
Growth diagnostics
Problem: Low levels of private investment and entrepreneurship
High cost of finance
Low return to economic activity
Low appropriability
Low social returns
poor natural
resource
management
poor
geography
low
human
capital
government
failures
bad international bad local finance
finance
market
failures
information
externalities:
“self discovery”
bad infra
structure
- micro risks:
macro risks:
property rights,
financial,
corruption,
monetary, fiscal
taxes
instability
-
coordination
externalities
low
domestic
saving
poor
inter
mediation
-
Source: Hausmann, Rodrik, Velasco (2005)
Why were bank deposit rates and risk premiums
high?
Were they high because of bad international finance?

International finance was good



Mongolia’s official debt was primarily concessional, and
long-term
FDI inflows were strong at the time of analysis
Outlook was also good



The spread on ‘B+’ Fitch rated countries was 280 to 300
basis points
Collateral could be used to bring down the spread
further down
The outlook has changed since then due to the sudden
negative TOT shock
Growth diagnostics
Problem: Low levels of private investment and entrepreneurship
High cost of finance
Low return to economic activity
Low appropriability
Low social returns
poor natural
resource
management
poor
geography
low
human
capital
government
failures
bad international
finance
market
failures
information
externalities:
“self discovery”
bad infra
structure
- micro risks:
macro risks:
property rights,
financial,
corruption,
monetary, fiscal
taxes
instability
bad local finance
-
coordination
externalities
low
domestic
saving
poor
inter
mediation
-
Source: Hausmann, Rodrik, Velasco (2005)
Why were bank deposit rates and risk premiums
high?
Was bad local finance the reason for the high cost of capital?


Domestic saving were rising in Mongolia
due to strong growth and BOP position
Rising official reserves and commercial
bank assets pushed the 2006 liquidity
ratio to 600% and credit growth was
highest since 1992
Growth diagnostics
Problem: Low levels of private investment and entrepreneurship
High cost of finance
Low return to economic activity
Low appropriability
Low social returns
poor natural
resource
management
poor
geography
low
human
capital
government
failures
bad international
finance
market
failures
information
externalities:
“self discovery”
bad infra
structure
- micro risks:
macro risks:
property rights,
financial,
corruption,
monetary, fiscal
taxes
instability
bad local finance
-
coordination
externalities
low
domestic
saving
poor
inter
mediation
-
Source: Hausmann, Rodrik, Velasco (2005)
Why were bank deposit rates and risk premiums high?
Poor financial intermediation was responsible for the high cost of capital


Bank deposit rates were high due to intensive
competition among financial institutions in
Mongolia
Spreads were high due to a combinations of
factors:



Difficulty in assessing credit risk;
High bank operating costs;
Low profitability of banks’ non-lending assets;
Were the high cost of capital and limited access to
capital the reasons for the large number of firms without
loans in Mongolia?
Loan
Maturity of
1 year
27%
All firms
100%
With a loan
Without a loan
27.9%
Loan
Maturity >
5 years
0.9%
72.1%
Applied
Did not apply
4.0%
Why?
Lack
Collateral
3%l
Low return
To capital?
1%
68.1%
Discouraged
25.9%
42.2
Why?
High cost of
Capital
22.0%
Source: Ianchovichina and Gooptu (2007)
Did not need a
loan
Why?
Collateral
18.7%
Low return
To capital?
42.2%
Discrepancy between subjective and
objective data in Mongolia

Cost of capital



Whereas 56% of the firms in the ICA complained that the cost of
capital is a severe obstacle to business growth
Only 22% of the firms in the survey did not apply for a loan because
of the high cost of capital
Access to capital



Whereas 42% of firms claim that access to credit was a severe
obstacle
70% either obtained a loan (28% of firms) or did not need a loan
(42% of firms)
Access to long-term financing is limited


Collateral requirement is excessive due to problems with assessing
credit risk
Conclusion: while the cost of capital was high, it was not the
primary reason for the small number of firms with loans
The case of Benin
Was Private Investment Low in Benin?





Gross domestic investment has been low by international
standards, averaging 18.2% of GDP in the last 10 years
Nearly all of private foreign investment was FDI, averaging just
1.7% of GDP in the past decade
 In line with WAEMU, but much below SSA, HIPCs and
LICs averages
Private domestic investment was a smaller share of domestic
investment than the average in WEAMU, HIPC, LICs
Only a small share of firms had loans in 2004 and most of the
loans to the private sector were short- to medium-term
As in other HIPCs a large share of investment in Benin was
funded by foreign aid
Were there signs that credit to the private
sector was tight?


Broad money rose by more than 22 percent in
2005, considerably higher than nominal GDP
No signs of crowding out


There was a 20 percent expansion of credit to the
private sector.
As net bank credit to the government has declined,
growth of credit to the private sector has remained at
nearly 10 percent, with some shift towards longer term
credit and lending to non-trade services, especially
telecommunications.
Benin Monetary Developments
(CFA Francs billions)
Source: International Monetary Fund, Article IV Consultation, January 2007
Was the low level of private investment in Benin a
signal of low supply or low demand for finance?
Real average cost of capital
15
10
2005
Source: SIMA and Government of Benin.
2006
Developing EAS
HIPC
ECOWAS other than
WAEMU
Egypt
Honduras
Thailand
Malaysia
Indonesia
China
Bangladesh
India
Mauritius
South Africa
Tanzania
-5
Nigeria
0
Kenya
5
Madagascar

20
Uganda

Average real cost of
capital was lower
compared to other
developing countries
It has risen during the
past 3 years
But in 2007, the cost of
capital for small
enterprises was close
to 8 percent – much
lower than the rates
faced by SMEs in
many developing
countries
Benin

Benin did not appear to be finance constrained…
Real interest rate
14
12
10
8
6
Mongolia
Source: Ricardo Hausman, “A framework for Growth Diagnostics”, Kennedy School of Government, Harvard University, May 2006.
Romania
Poland
Ecuador
Pakistan
Bosnia
Nicaragua
Bangladesh
Algeria
Moldova
India
Philippines
Slovakia
El Salvador
Czech
Kenya
Morocco04
Estonia
Bulgaria
Croatia
Morocco00
Indonesia
Russia
Egypt
Hungary
Albania
China
Brazil
Slovenia
Turkey
4
2
0
-2
-4
-6
Oman
Benin
Cost of finance was low given the availability
of credit to the private sector…
(Domestic Credit to private Sector / GDP)*100
Domestic Credit to Private Sector (% of GDP) vs Real
Interest Rate. Constrained Countries 2000-2007
Mean Real Lending Rate = 8.4%
50
Mean Dom Cred to Priv / GDP = 49.5%
40
Mean Real Lending Rate + 1 std Dev = 15%
Brazil
30
Mongolia
20
Tajikistan
10
Benin
Zambia
(Mean Dom Cred to Priv / GDP) -1 Std Dev= 3.8%
0
0%
5%
10%
15%
20%
25%
30%
Real Lending Rates
35%
40%
45%
50%
Perceptions differed from reality in Benin


78 % of the firms
complained that the
cost of capital was a
severe obstacle to
business growth, but
only 10 % of the
firms were
discouraged and did
not apply for a loan
because of the high
cost of capital
70 percent of firms
claimed that access to
credit was a severe
obstacle, 60 percent
of the firms either
obtained a loan, were
approved for a loan or
did not need a loan.
Access to capital in Benin, 2004 (ICA)
Loan
Maturity of
1 year or less
16 (8.4%)
Loan
1<Maturity <=
5 years
26 (13.6%)
All firms
191 (100%)
With a loan
Without a loan
49
(25.7)
Applied
Did not apply
49
(25.7%)
Loan
Maturity >
5 years
7 (3.7%)
93
(48.7%
Discouraged
Why?
Rejected
15
(7.9%)
Other:
5 (2.6%)
142
(74.3%
Lack of
collatera
l: 8
(4.2%)
Source: Ianchovichina (2008) based on Benin
Investment Climate Survey 2004.
Project
not
feasible:
2 (1.1%)
Not
rejected 34
(17.8%)
High cost of
Capital
19 (9.9%)
65 (34%)
Collateral/
Insufficient
guarantee
8 (4.2%)
Process too
Difficult
22 (11.5%)
Did not need a loan
28 (14.7%)
Other
16
(8.4%)
Real cost of capital by type of
borrower in Benin
Nature of borrower
State or para-state organizations
Private individuals
Financial customers
State companies and EPIC
Retirement Insurance fund
Private companies in productive
sector
Small enterprises
Village cooperatives and groups
Other (NGO, Friends, Unions, etc.)
Personnel of banks
Total
Credit
Deposit
2005
1.5
5.8
2.4
2.3
4.1
3.5
7.5
5.9
5.3
-2.6
3.7
Credit
Deposit
2006
-1.1
-1.5
-0.3
-0.2
-0.4
-0.6
0.3
6.9
8.6
4.1
7.0
5.9
0.5
0.6
1.8
1.4
1.5
1.2
-1.5
-1.2
-1.3
-5.4
-0.7
8.1
6.2
5.1
-1.0
5.8
-0.1
1.5
-0.1
0.2
1.1
Credit
Deposit
2007
1.7
8.3
2.7
5.7
4.4
8.1
2.1
1.7
3.3
3.2
2.7
2.5
9.8
6.7
4.4
0.5
7.6
0.8
1.1
1.3
1.3
2.2
Source: SIMA and Government of Benin
Real cost of capital for small enterprises was rising and was much higher
than the average, but still lower than in many countries and access to
microfinance did not appear to be a problem
Microfinance in Benin is a dynamic sector





Microfinance has grown tremendously in the last decade
Benin has the largest number of microfinance institutions in the
WAEMU region.
In 2002, there were more than 600 retail microfinance
organizations belonging to about 85 programs or networks
reaching about 500,000 people
A penetration rate of about 15 percent of the total active
population
However, access to long term capital that can fund productive
investment, not short-term trade-related activities, is very
limited
The case of Zambia
Average cost of finance declined in recent
years
Figure 1: Real cost of capital (average)
100.0
90.0
80.0
Percent
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
1995
1996
1997
1998
Zambia
1999
2000
Mongolia
Source: Authors’ own calculations using World Bank data.
Ianchovichina and Lundstrom (2008)
2001
2002
South Africa
2003
Uganda
2004
2005
But, cost and access of capital
differentials were sizable

Access and cost of capital varied with firms’ size





In 2003, nearly 50 percent of larger firms had a loan, while only 19 percent
of small firms had a loan
The cost of capital differential between large and small firms was more
than 10 percentage points
Similar differentials existed between the cost of capital of exporters and
non-exporters, domestic and foreign companies
Micro firms faced even steeper constraints
Access and cost of capital varied by area


Rural areas had very limited access to capital
Access to capital through informal channels at prohibitively high cost
What were the reasons for the poor access to and high
cost of finance for small and micro firms?

Poor financial intermediation rather than low domestic savings or bad
international finance



Domestic savings as a share of GDP climbed up from 6% in 1990s to 18.1% in 2006,
a share higher than the SSA average
FDI and aid were higher than the average for SSA and LICs both in 1990s and 2000s
Financial intermediation was limited by small size of banking sector,
and an inadequate supporting financial infrastructure
Percentage of population with a bank deposit account
Ianchovichina and Lundstrom (2008)
Despite limited use, there were signs of improvement


Only 5 to 8 percent of business owners used microfinance (FinTrust 2007)
Signs of improvement:


the percentage of people identifying the cost of finance as the main reason for
their poverty status halved in the period 2002-06
Some micro finance institutions operated by NGOs and outgrowers schemes
successful in providing credit to farmers but limited coverage
Number of commercial banks' branches
185
180
175
170
165
160
155
150
145
140
2001
2002
2003
2004
2005
Source: Bank of Zambia (2007).
2006
2007
Top reason for not using financial service
was lack of income
Reasons for not having a bank account
Source: FinTrust (2007). FinScope data from 2005.