Overseas Development

Institute

Global Financial Crisis

Discussion Series
Paper 23: Methodological note:
update for phase 2 studies

Global Financial Crisis Discussion Series

Paper 23: Methodological Note:
Update for Phase 2 Studies1

February 2010

Overseas Development Institute

111 Westminster Bridge Road
London SE1 7JD
www.odi.org.uk

1 This paper is part of a wider research project coordinated by the Overseas Development Institute (ODI) London, led by Dirk
Willem te Velde and including ODI (Isabella Massa, Massimiliano Cal, Jodie Keane, Nicola Cantore and Sheila Page) and other
researchers, and supported by the UK Department for International Development (DFID) and the Swedish International
Development Agency (Sida), but it does not necessarily reflect their views.

Contents
Figures, tables and boxes
Acknowledgements
Acronyms
Abstract

iii
iv
v
vi

1.
2.

1
3
3
6
7
8
9
10
10
10
13
14
16
16
16
17
17
17
19
19

Introduction
Effects of the global financial crisis: Key transmission mechanisms
2.1 Private capital flows
2.2 Trade
2.3 Remittances
2.4 Aid
2.5 Summary: BoP effects
3. Growth and development effects
3.1 National-level growth, investment and employment
3.2 Fiscal effects
3.3 Sectoral-level effects
3.4 Poverty and distributional effects
4. Policy responses: A critical review
4.1 Macroeconomic policies to manage the impact of the crisis
4.2 Social policies to respond to the impact of the crisis
4.3 Economy-wide and sectoral structural policies for getting the country out of the crisis
4.4 Institutional context and constraints of policymaking
4.5 Multilateral and bilateral donor responses in-country
5. Conclusions and looking ahead
5.1 The impact of the crisis: An update
5.2 Looking ahead: How well is the country positioned to gain from a future recovery
and to grow sustainably?
References
Annex 1: Commodity prices

ii

19
20
21

Figures, tables and boxes

Figure 1: Mapping out the effects of the global crisis and policy responses
Figure 2: International reserves in DRC, Jan 2008-Mar 2009 (in US$ end of the month)
Figure 3: Cambodian real GDP growth, actual and forecasts, 2002-2009 (%)
Figure 4: Trends in budget deficit in Bangladesh, FY2000/01-FY2008/09
Figure A1: Price developments of crude oil (US$/bbl)
Figure A2: Price developments of aluminium (US$/mt)
Figure A3: Price developments of gold (US$/toz)
Figure A4: Price developments of copper (US$/mt)

2
9
10
12
21
21
21
21

Table 1: Key questions to understand financial vulnerability and financial contagion in the
developing world
Table 2: FDI inflows in study countries, 2004-2007 (US$ millions)
Table 3: FDI inflows in selected study countries, 2007 (% of GDP)
Table 4: Selected sub-Saharan African countries share of banking assets held by foreign
banks with majority ownership, 2006 (%)
Table 5: Total international banks claims in study countries, Mar 2007-Mar 2009
Table 6: Trade issues, indicators and data sources
Table 7: Measuring the effects of the crisis on migration and remittances
Table 8: Remittances in study countries, 2004-2008
Table 9: Official development assistance (ODA), 2008 (% of GDP)
Table 10: Distribution of mining sector tax revenues in Zambia, 2008 (kwacha billions)
Table 11: Expected poverty effects of global financial crisis in 2009

5
6
6
8
8
8
12
14

Box 1: Measuring remittances

Box 2: Commodity dependence and budgeting (Sudan, Bolivia and Zambia)

8
13

iii

4
5
5

Acknowledgements
A team of 40 researchers from ODI and 10 developing countries embarked on a unique monitoring
study on the effects of the global financial crisis over the period January-March 2009. They found that
some effects had already become visible when the G-20 leaders met in London on 2 April 2009, but
worse conditions were thought to be underway. The monitoring work is now being updated and new
countries are being included, to total 11. This paper provides a methodology to enable country authors
to update their country case studies.
We are grateful to the UK Department for International Development (DFID) and the Swedish
International Development Cooperation Agency (Sida) for their support for Phase 2 of the monitoring
work.

iv

Acronyms
ADB
AGOA
ASEA
BIS
BoP
CAR
CGE
DFID
DRC
EAP
EIU
EU
FDI
GDF
GDP
GNI
IDA
IFC
IFI
IFS
ILO
IMF
IPO
NPL
ODA
ODI
P/E
R&D
RDB
ROA
ROE
SAS
SDR
Sida
SSA
TPR
UK
UN
UNCTAD
US
USITC
VAT
WDI
WEO
WTO

Asian Development Bank

African Growth and Opportunity Act
African Securities Exchanges Association
Bank for International Settlements
Balance of Payments
Capital Adequacy Ratio
Computable General Equilibrium
UK Department for International Development
Democratic Republic of Congo
East Asia and Pacific
Economist Intelligence Unit
European Union
Foreign Direct Investment
Global Development Finance (World Bank)
Gross Domestic Product
Gross National Income
International Development Association
International Finance Corporation
International Financial Institution
International Financial Statistics (IMF)
International Labour Organization
International Monetary Fund
Initial Public Offering
Non-Performing Loan
Official Development Assistance
Overseas Development Institute
Price to Earnings
Research and Development
Regional Development Bank
Return on Assets
Return on Equity
South Asia
Special Drawing Right s
Swedish International Development Agency
Sub-Saharan Africa
Trade Policy Review (WTO)
United Kingdom
United Nations
UN Conference on Trade and Development
United States
US International Trade Commission
Value Added Tax
World Development Indicators (World Bank)
World Economic Outlook (IMF)
World Trade Organization

Abstract
This paper provides an update of the methodology that country case studies can use when undertaking
their research.
The methodology discusses how to report on:
The national or regional components of the shock, to determine which countries are linked
most closely to centres of the shock;
What the effects on growth, investment, sectors, poverty and inequality are;
What the policy responses have been.
Transmission mechanisms include:
Private capital flows;
Trade;
Remittances;
Aid;
Summary: balance of payments and exchange rate effects.
Growth and development effects cover:
National-level growth effects;
Fiscal effects;
Sector effects;
Poverty and inequality.
Policy responses can be divided into:
Macroeconomic policies;
Social policies;
Economy-wide and structural policies;
Donor responses.
Finally, we need to look ahead and see whether the country is well placed to gain from a recovery and
to respond effectively to crises in the future.

vi

1. Introduction
There is now broad agreement that the global financial crisis will have major impacts on developing
countries, but that these will vary markedly. Phase 1 research, coordinated by the Overseas
Development Institute (ODI) and involving developing country teams in 10 countries, showed that,
while the transmission mechanisms are similar (trade, private capital flows, remittances, aid), the
effects vary by country and therefore country-specific monitoring is required. Most findings suggest
that, as a result of time lags, the worst of the effects is yet to come.
Phase 2 case studies will aim to:
Update information on the effects of the global financial crisis, based on the methodology
developed in Phase 1, by examining the following transmission belts:
o Private financial flows (splitting out portfolio flows, foreign direct investment (FDI) and
international bank lending);
o Trade (imports and exports);
o Remittances; and
o Aid.
Update information on economic (growth) and social (employment and poverty) effects.
Monitor policy responses.
Deepen the existing analysis where feasible, e.g. on effects on special groups or industries.
Discuss country-specific issues.
This paper is a methodological note which can be used for Phase 2 studies.
Figure 1 illustrates the main building blocks of the Phase 1 and 2 methodology:
What is the global shock (while effects are still getting worse, some agreement on type and
order of magnitude of shocks)?
The national or regional components of the shock, to determine which countries are most
closely linked to centres of the shock.
What are the effects on growth, investment, sectors, poverty and inequality?
What have been the policy responses?

Figure 1: Mapping out the effects of the global crisis and policy responses

What is the shock at global level?

Global shock

Level and type of economic and financial integration

Characteristics and distribution of links with other countries
Structure of economy
Policy

What are the components of the shock at national level?

Trade

Private capital flows

Remittances

Aid

Economic/social structures
Institutions/policies
Assets, prices, employment
Access to goods and services

What are the broad macro effects? (effects so far, possible effects)

Inequality,
poverty

Growth, investment and employment

National and sectoral

Public and
private debt

Critical review of policy responses

Macroeconomic policies to
manage shock (fiscal,
monetary, financial)

Social policies to
manage shocks

Accelerating normal
growth and
development policies

Is the country well placed to benefit from a recovery, and can it respond more effectively to
future crisis?

2. Effects of the global financial crisis: Key transmission

mechanisms
Before we move on to the specifics of the transmission mechanisms, we need to acknowledge the
difficulty in isolating the effect of the global financial crisis when there are other shocks happening.
Here we suggest possible tools that might be used to understand and describe what is going on (note
that the emphasis of the study lies on monitoring and not on detailed modelling):
Before/after comparisons: For instance, what were the trends before July 2008 and what
happened afterwards? Although this is crude, it could be a first guide, especially if no formal
models are readily available. For instance, have tourist bookings, exports or FDI approvals
suddenly declined, without any other country-specific shock or event occurring, such that the
changes can, at a first analysis, be attributed to the global financial crisis?
Partial equilibrium models, used implicitly or explicitly: These take into account other factors
(i.e. construct a counterfactual on what would have happened had there been no crisis).
o On FDI, one could use existing partial equilibrium models, such as Inward FDI = f
(GDPhome, GDPhost, GDPgrowthhost, risk, etc). Various existing bilateral FDI models exist
which e.g. suggest that a 1% decline in home (and host) country gross domestic product
(GDP) reduces the stock of FDI to developing countries by around 0.2% (and 0.5%). Such
models may also exist for specific countries. Has this happened?
o There will be studies on export supply/demand by country with published price and income
elasticities, which can be employed to examine possible effects, because we can use
estimates of possible drops in export markets and recent decreases in commodity prices. It
may also be possible to use existing computable general equilibrium (CGE) models.
o On remittances, there are models such as REMIT = f (GDPhome, GDPhost, emigration rate),
which some studies have estimated.
o On aid, there is no need for a model because aid is largely a policy not a behavioural
variable, such as private sector exports.
For all these, studies then need to identify what part of any change in international variables
owes to the shock and what part owes to other influences.
CGE models: While these may not be the best tools to identify shocks, they are useful to
simulate effects of shocks on the rest of the economy.
Interviews with public and private sectors about perceived impact. For example, it can be very
helpful to discuss impacts with banks because they have a cross-sector overview of the latest
impacts.

2.1

Private capital flows

In order to assess the vulnerability of a specific country to a shock in private capital inflows, the
analysis could focus on a number of factors. It is important to understand the countrys dependence on
FDI, portfolio flows and international bank lending. It would also be helpful to identify sectors and
investor countries. With respect to international bank lending, it could be worth investigating degree of
exposure of banks to foreign ownership as well as the countries of origin of foreign banks.
Table 1 provides a series of key questions for examining developing countries financial systems and
determining whether developing countries are at risk of financial contagion.

Table 1: Key questions to understand financial vulnerability and financial contagion in the
developing world
Questions

How sound is the domestic

banking system?

How vulnerable is the domestic

banking system to the financial
crisis?

How strong are the links among

stock markets internationally?
Is there stock market financial
contagion between developing
and developed countries?
Are investors losing confidence
in developing countries?
Has investor risk aversion
towards developing countries
increased?
Has the financial turmoil
tightened credit conditions to
developing countries?

Indicators
Risk-weighted capital adequacy
ratio (CAR)
Return on assets and equity (ROA
and ROE)
Non-performing loans (NPLs)
Banks ownership structure
Growth of bank credit to private
sector
Composition of bank lending
Securities investment to total bank
assets
Exposure to new financial
instruments
Total foreign holdings of equity

Data source suggestions

International Monetary Fund (IMF)
Global Financial Stability Report,
national sources

Share prices correlation

Equity prices trend
Price-to-earnings (P/E) trend
Equity portfolio flows
Number of initial public offerings
(IPOs) withdrawn or postponed
Equity issuance
Bond issuance

Bloomberg, Datastream, national

stock exchanges

Sovereign and/or corporate bond

spread
Growth of bank lending rate
New loans issuance

Bloomberg, IMF IFS, Reuters

DealScan database (for large
banks loans), national sources

World Bank Global Development

Finance (GDF), IMF International
Financial Statistics (IFS), national
sources

IMF World Economic Outlook (WEO)

and IFS

National sources, World Bank

World Development Indicators
(WDI)
World Bank GDF, IMF WEO, national
sources

At this stage of the monitoring work, it is important to report the magnitude of the country-specific
shock on the basis of what is already visible or published with respect to the three different types of
private capital flows:
FDI inflows: Report on developments using balance of payments (BoP) data from central banks
and planned investment data from investment promotion agencies; as background information,
the United Nations Conference on Trade and Development (UNCTAD) FDI On-line Database, the
IMF IFS and the World Bank WDI can be useful.
Portfolio investment flows (portfolio equity flows and bond flows): Report on withdrawals from
stock markets, trends in main share price indexes, stock market capitalisation, stocks traded
value, turnover ratio, equity issuance, IPO activities and bond issuance, using data from
national stock exchanges, regional stock exchange associations (e.g. the African Securities
Exchanges Association ASEA) and other national sources;
International bank lending: Report on developments using the Bank for International
Settlements (BIS) Consolidated Banking Statistics, the World Banks GDF and national sources.
Table 5 shows there have been large declines in bank lending to African countries up to
2009Q1, but what has happened since?

Table 2: FDI inflows in study countries, 2004-2007 (US$ millions)

Ethiopia
Kenya
Mozambique
Uganda
Tanzania
Zambia
Democratic Republic of Congo (DRC)
Sudan
Bolivia
Bangladesh
Cambodia
Source: UNCTAD FDI On-line database.

2004
545.1
46.1
244.7
295.4
330.6
364.0
9.9
1511.1
85.5
460.4
131.4

2005
265.1
21.3
107.9
379.8
567.9
356.9
-76.0
2304.6
-287.8
845.3
381.2

2006
545.3
50.7
153.7
400.2
521.7
615.8
-116.0
3541.4
280.8
792.5
483.2

2007
254.1
728.0
427.4
367.9
599.5
983.9
720.0
2436.3
204.2
666.4
867.3

Table 3: FDI inflows in selected study countries, 2007 (% of GDP)

2007
Ethiopia
1.5
Kenya
2.4
Mozambique
5.1
Uganda
3.0
Tanzania
4.0
Zambia
8.7
Sudan
5.4
Bolivia
1.7
Bangladesh
0.9
Cambodia
11.1
Source: UNCTAD FDI On-line database.

Table 4: Selected sub-Saharan African countries share of banking assets held by foreign banks
with majority ownership, 2006 (%)
Togo
Malawi
Burundi
Congo
Senegal
Angola
Mali
Niger
Burkina Faso
Tanzania
Rwanda
Zambia
Uganda
Mozambique
Madagascar
Source: World Bank GDF 2008.

2006
0
22
36
47
48
53
57
59
65
66
70
77
80
100
100

Table 5: Total international banks claims in study countries, Mar 2007-Mar 2009
Mar 2007 (US$m)

Mar 2008 (US$m)

Bangladesh
1567
1843
Bolivia
379
430
Cambodia
69
123
DRC
440
495
Ethiopia
62
56
Kenya
1347
1651
Mozambique
688
1324
Sudan
1032
1294
Tanzania
660
775
Uganda
287
712
Zambia
374
1385
Source: BIS Consolidated Banking Statistics 2009.

2.2

Mar 2009 (US$m)

2073
339
225
532
28
1656
1138
404
972
467
1127

% change Mar
2008-Mar 2009
12.5
-21.2
82.9
7.5
-50.0
0.3
-14.0
-68.8
25.4
-34.4
-18.6

Trade

Table 6 provides a checklist of issues important for examining the effects of the crisis on trade, trade
patterns and trade prices. Key questions include:
What have been the recent developments in trade values, exports and imports of goods and
services, by country and key products, for the latest month or quarter?
What is the experience on import and export prices? Use IMF data, which are more common and
recent than World Bank data (reported in the Annex). Assess impact of trade prices on declines
in values of exports.
What are the countrys main export products? How concentrated/diversified are the main export
products and markets?
What are the main products that it imports? How concentrated are these, and from which
countries do they come?
Have recent fluctuations been experienced previously, and have similar shocks been
experienced?
Table 6: Trade issues, indicators and data sources
Issue
Sectoral composition of
economy

Dependence on trade;
openness to trade; trade
performance
Recent developments in
main export products and
main destination markets

Main import products and

main import markets

Indicators
Contribution of agriculture, mining,
manufacturing, services to GDP
Main manufacturing/service activities
contribution to GDP
Exports as % of GDP, imports as % of
gross national income (GNI)
Current account balance (in current
US$/national currency and as % of GDP)
10 major exports (per export value), on a
HS-6 digit level, 2003-2008 (in volumes
and value) and/or indices of
concentration
Same for imports
2008 and 2009 monthly values and
volumes
10 major imports (per import value), on a
HS-6 digit level, 2003-2008 (values only)
2008 monthly values and volumes for 10
major imports (in national currency)

Data source
ODI: WDI, World Trade
Organization (WTO) Trade
Policy Review (TPR)
ODI: WDI, WTO TPR

ODI: Comtrade; Eurostat;

United States International
Trade Commission (USITC)
Interactive Tariff and Trade
DataWeb; other bilateral trade
statistics (as far as available)
Country experts: national
trade statistics
ODI: Comtrade; Eurostat;
USITC Interactive Tariff and
Trade DataWeb; other bilateral
trade statistics (as far as
available)
Country experts: National
trade statistics

Terms of trade

Assessing options and

constraints of market
diversification

Projections for major

commodities

Is the lack of liquidity

affecting exports?

Trade in services: tourism

2.3

Indices of nominal and price deflated

exchange rates (national currency vs.
currencies of main export markets);
2000-2007; 2007-2008 monthly basis
Exchange rate developments of national
currency compared with currencies of
main import markets; 2000-2007; 20072008 monthly basis
Market access regimes in main export
markets (e.g. duty-free quota-free in
US/EU (European Union); rules of origin
for Cambodias apparels). Actual and
prospective changes in these (proposed
reforms in coverage of US African Growth
and Opportunity Act (AGOA) for example;
new schemes by India, China, etc)
Unit price received for 10 main exports
compared with world market prices
Predictions of global demand and price
developments for copper (Zambia team);
oil (Nigeria team); flowers in the EU
(Kenya team); cocoa in the EU market
(Ghana team); garments in the US
(Cambodia team); tourism (all)
Access to trade finance

Tourist arrivals/registrations
Bookings/cancellations
Expenditure per tourist

ODI: IMF IFS

ODI: WTO TPR; Ministry of

Trade and Industry; Comtrade
Country experts: firm
interviews what are
constraints of market
diversification for major
exports?

ODI: World Bank and other

international sources on
commodity projections
Country teams: firm
interviews; export
associations
ODI: e.g. banks SWIFT
payment systems
Country experts: company
interviews how have the
costs of trade financing
increased (e.g. letters of
credit)?
ODI: WDI; World Travel and
Tourism Council
Country experts: central
banks; Ministry of Interior
(Cambodia); tourism board;
hotel associations

Remittances

Remittances to developing countries have been growing, but it is likely that they will fall in 2009 in
many case study countries. Relevant data could include:
How much are remittances affected, by country of origin check with central bank on BoP
data.
How much have migration flows and migration stocks been affected visitor and migration
statistics (this includes return migration).
How dependent are countries on remittances and how has this dependence evolved over the
past few years? Based on World Bank data.
What do the money transfer institutions report? What is the accuracy of official estimates, and
what share of remittances is unrecorded?
Table 7 provides suggestions on how country case studies can examine the effects of the crisis on
remittances and migration flows. Box 1 discusses how remittances are normally measured. An analysis
of the composition of the migrant population by country of destination and by socio-demographic
characteristics would be useful to make inferences on the extent to which migration and remittances
might be affected in the future.

Table 7: Measuring the effects of the crisis on migration and remittances

Question
What impact
on migration?

Source
Professional associations

Recruitment agencies

What impact
on
remittances?

Docquier et al. (2007)

(analysis by ODI)
BoP

Parsons et al. (analysis by

ODI, Cali and DErba, 2009)

Possible indicator
1. No. of departures of professionals per month
2. No of returnees per month (to be compared with same
period in 2007 and 2006)
3. No. recruited per 100 candidates (monthly data to be
compared with same period in 2007 and 2006)
4. Composition of migrant population by skills and gender
1. Worker remittances + compensation of employees + migrant
transfers (monthly data to be compared with same period in
2007 and 2006)
2. Composition of migrants population by country of
destination

Box 1: Measuring remittances

1) Workers remittances recorded under the heading current transfers in the current account (item code 2391 in
the IMFs BoP Yearbook)
2) Compensation of employees, which includes wages, salaries and other benefits of border, seasonal and other
non-resident workers (such as local staff of embassies) and which are recorded under the income subcategory of
the current account (item code 2310)
3) Migrants transfers, which are reported under capital transfers in the capital account (item code 2431)

Table 8: Remittances in study countries, 2004-2008

2004
2005
2006
2007
2008e
2007 share in
(US$m)
(US$m)
(US$m)
(US$m)
(US$m)
GDP (%)
Bangladesh
SAS
3584
4314
5428
6562
8893
9.5
Cambodia
EAP
177
200
297
353
353
4.2
Ghana
SSA
82
99
105
117
128
0.8
Kenya
SSA
620
805
1128
1588
1673
5.4
Nigeria
SSA
2273
3329
5435
9221
9979
6.7
Zambia
SSA
48
53
58
59
59
0.5
Notes: e. Estimates based on data until October 2008. SAS = South Asia; EAP = East Asia and Pacific; SSA = SubSaharan Africa.
Source: World Bank based on IMF BoP statistics.
Region

2.4

Aid

Here it will important to report:

Volume of aid (in US$);
Development in aid dependence (% of GDP), filling out table below;
Proposals for changes in aid by each countrys major donors (this could be mentioned in
Section 3 instead).
This is based on annual data, although there might be more recent evidence for shifts in aid flows.
More discussion on aid can be found in the policy response section.
Table 9: Official development assistance (ODA), 2008 (% of GDP)
Country
IMF (2009)
WDI 2005
Bangladesh
1.2
2.1
Bolivia
1.2
6.5
Cambodia
6.4
9.1
Kenya
0.8
4.1
Uganda
4.9
14.0
Zambia
4.5
13.9
Source: IMF (2009), WDI and Phase 1 country reports

Phase 1 country report

1.6% of GDP (2007)
9% of GDP in 2007
4 % of GDP (2006)
13.2% of GDP in 2006

2.5

Summary: BoP effects

This section could summarise the BoP of the shocks: how do all the transmission belts come together?
It could report on:
Summary BoP (from central bank statistics), quarterly where feasible.
Developments in international reserves (expressed in terms of months of imports, e.g. from
IMF). See Figure 2 for an example.
Exchange rate developments Has the change in capital and current account flows put pressure
on the exchange rate? Use central bank statistics for nominal and IMF for real effective
exchange rates.
Figure 2: International reserves in DRC, Jan 2008-Mar 2009 (in US$ end of the month)
300
270.7

250

252.8

249.9

237.3

200

150

210.7
181.2

185.7

181.8

219.7
189.5

184.7

159.0

100

34.5

32.9

Jan-09

Feb-09

78.6

50

Source: Kabuya and Cassimon (2009), presentation on 7 September 2009. See also te Velde (2009)

Mar-09

Dec-08

Nov-08

Oct-08

Sep-08

Aug-08

Jul-08

Jun-08

May-08

Apr-08

Mar-08

Feb-08

Jan-08

3. Growth and development effects

3.1

National-level growth, investment and employment

The following points are important:

It is key to highlight current and past real GDP growth rates and, where possible, changes in
growth rates of these components of GDP (including investment, consumption and
employment).
Some countries will be able to report quarterly GDP statistics be clear if you are using quarteron-quarter growth rates (annualised) or whether it is, say, 2009Q2 over 2008Q2 annual growth
rates.
Use calendar years, converting as necessary from any fiscal years.
Use growth forecasts by national and international (IMF) sources and, where feasible, compare.
The example below is quite helpful.
Beyond this, we think it is possible to have a growth narrative. Why did growth fall (e.g. Cambodia) or
remain fast (Zambia, Uganda)? Which transmission channel contributed most?
Figure 3: Cambodian real GDP growth, actual and forecasts, 2002-2009 (%)
15
13.3
10.8

10.3

10

10.2
6

8.5
6.6

4.9

5
4.7

- 0.5
2002

-5

4.8

2003

2004

IMF (old proj.)

ADB

2005

2006

IMF (new proj.)

Government

2007

2008e

2009p

World Bank
EIU

Notes: ADB = Asian Development Bank; EIU = Economist Intelligence Unit.

Source: Cambodia case study (Phase 1).

3.2

Fiscal effects

The effects of the global financial crisis work through a number of mechanisms, one of which is
government budgets. From a static point of view, GDP (Y) is determined by consumption (C) and
investment (I), from government sources (G) and private sources (P) plus net exports, exports (X) imports (M):
Y = PC + PI + GC + GI + X - M
Thus, growth is affected directly by total spending by government GS=GC+GI.
The government balance (GB) depends on spending (GS) and revenues (GR)
GB = GS --GR -- INTPAY
10

where INTPAY is interest payments on government debt. If government/fiscal balance is negative it

adds to government debt (GD), via (INTPAY=r*GD).
GD = GD (-1) + GB
If GB is negative for too long, the continuing increase in GD will raise doubts about government debt
sustainability.
Government revenues depend on tax revenues (corporate tax, income tax, trade taxes, other taxes and
royalties), fees for government services, capital income and grants.
GR is likely to decline in the global financial crisis because:
Corporate taxes decline with lower private activity;
Royalties and mining taxes decline as commodity exports in volumes and values decline;
Import taxes decline because import values decline with lower commodity prices and lower
import volumes because of lower total final expenditure (TFE=Y+M);
Capital income (profits) is lower;
Grants (e.g. aid) are lower.
Authors of studies are asked to examine whether the above decline has occurred, with the result being
a higher fiscal deficit or lower surplus. The IMF suggests that many African countries have increased
their fiscal deficit by taking in fewer import taxes.
Authors can also look for positive developments. For instance, GS could be lower, e.g. for oil importers
when oil prices are lower.
There can also be reallocations in the government budget, from investment to consumption, or shifts
within these categories.
It may also be important to think about the funding part of government debt: have international bond
markets eased up, and are countries able to raise finance in this way and what are the spreads (e.g.
compared to US Treasury Bills).
Government taxes and spending can also have other static effects on growth. If the government
reduces its tax receipts owing to the above effects, this should affect PC (through real disposable
income) and PI (through the user cost of capital, which depends on taxes and interest rates).
Government taxes and spending can also have dynamic effects on growth. Tax incentives can be linked
to behaviour of investors (e.g. rewarding research and development (R&D) investment or skills training)
and spending can have externalities/multiplier effects on growth.
3.2.1 Data availability
Budget plans and outturn: these tend to be annual data, although in some cases monthly and quarterly
data are available.
3.2.2 Examples from Phase 1 country case studies
Uganda uses quarterly data suggesting that tax revenues were lower in 2008/09Q3. This owed
especially to lower domestic tax revenues, as value added tax (VAT) revenues were lower because of
lower domestic consumption.
Bangladesh records savings on budgetary allocations for fuel subsidies. Monthly data show that
collection of import duties experienced a sharp fall in the months at the end of 2008, with only
marginal improvement in the month of January 2009. Import duty and other import-related duties

11

together constitute around 42% of total National Board of Revenue collection in Bangladesh. Thus, it is
likely that revenue collection will fall short of the target.
Figure 4: Trends in budget deficit in Bangladesh, FY2000/01-FY2008/09
180000
4.9

140000

4.4

5.0

crore Taka

120000

3.5

3.4

3.5

100000

3.0

80000

3
2.8

Total Expenditure

60000

40000
20000
0
FY01

Total Revenue
FY02

FY03

per cent

160000

6
5.2

FY04

Total Revenue

FY05

FY06

Total Expenditure

FY07

FY08

0
FY09

Deficit as % of GDP

Source: Ministry of Finance (Bangladesh case study Phase 1).

Net budgetary gains (e.g. lower fertiliser expenditure owing to lower import prices) could create some
fiscal space, allowing the government to go for higher expenditure without overshooting the projected
deficit of 5% of GDP.
In Zambia, the global financial crisis has come at a time when the mines are already contributing less
to government revenue than they were several years ago, although over the past few years there has
been some pickup and the forecast (before the crisis) was that the sector would contribute significantly
to tax receipts. The sector contributes 10% of total tax revenue and 1.8% of GDP. A loss of potential
revenue from the mines as a result of the crisis reduces the governments fiscal space to finance social
sector expenditure programmes such as education, health and infrastructure for poverty reduction.
Table 10 shows tax revenues for 2008. Countries could report using quarterly data, particularly when
comparisons can be made over more than one year.
Table 10: Distribution of mining sector tax revenues in Zambia, 2008 (kwacha billions)
Total tax revenue
Mining company tax

Jan-Mar
2113.6
116.1

Apr-Jun
2402
154.6

Jul-Sep
2667.3
146.8

Oct-Dec
2495.2
24.2

Windfall tax
Mineral royalty

18.2

85.7

118.3
95.8

7.7
59.3

Export duty on copper

concentrates
Total mines contribution

19.1

23.1

70.8

64.9

263.4
(11.0)

431.7
(16.2)

156.1
(6.3)

153.4
(7.3)*
Note: * as a percentage of total revenue.
Source: Zambia Revenue Authority.

12

2008
9678.1
441.7
(4.6)*
126.0
259.1
(2.7)
178.1
(1.8)
1004.9
(10.4)

Box 2: Commodity dependence and budgeting (Sudan, Bolivia and Zambia)

The global financial crisis meeting on 7 September at ODI suggested that it could be important to compare
budgeting practices across countries whose budgets rely on commodity prices (e.g. Sudan, Bolivia and perhaps
Zambia, which is also looking into mining, as is DRC), with respect to the impact of the global financial crisis.
Countries can use a common framework to understand the challenges.
What is the share of commodity taxes in total government revenues? Report quarterly or monthly data. How has
the fiscal position changed recently?
Was the budget based on high oil and commodity prices? What values? And was the country prepared for
commodity price shocks?
What was the result in terms of budget cuts and stimuli?

3.3

Sectoral-level effects

Here, we are interested in statistics on sectors, e.g. how have garments exports, tourism, construction,
commodities, etc, in terms of GDP, investment, employment, been affected?
In some countries, the effects are localised in areas or sectors, e.g. commodity exporters such as
Zambia. In others, it is a range of sectors (Cambodia).
Different countries could examine different sectors, and compare countries with similar shocks to
explain differences and commonalities.
3.3.1 Garments in Bangladesh and Cambodia
It will be useful to further compare the impact of the crisis on the garment sector in Cambodia and
Bangladesh. Countries could work with a common framework to understand the challenges.
What have been the effects so far on the garment sector, using relevant statistics? Output,
employment and foreign earning losses, etc. Report quarterly data or monthly, or anecdotal
evidence.
What are the main value chains in which garment companies operate (and does nationality
matter?) Where do they export to?
How competitive is the sector? What is the underlying cost structure (transport costs, cost of
capital, etc)?
How well is the sector positioned to bounce back (and what are key factors behind this
infrastructure, business environment, ability to introduce new polices)?
3.3.2 Flowers in East Africa (a similar study could be done for coffee)
It may be possible to compare the impact of the crisis on the flower sector across East Africa (e.g.
Uganda, Ethiopia, Kenya). Countries could work with a common framework to understand the
challenges.
What have been the effects so far on the flower sector, using relevant statistics? Output,
employment and foreign earning losses, etc. Report quarterly data or monthly, or anecdotal
evidence.
What are the main flower companies (and does nationality matter)? Where do they export to?
How competitive is the sector? What is the underlying cost structure (transport costs, cost of
capital, etc)?
How well is the sector positioned to bounce back (and what are key factors behind this
infrastructure, business environment, ability to introduce new polices)?

13

3.4

Poverty and distributional effects

The financial crisis affects four key types of resource flow to developing countries, including trade,
international capital flows, remittances and aid. These in turn will affect firms and households through
a number of key transmission channels, including:
Taxes and transfers: Public and private transfers and taxation, including targeted transfers,
subsidies, taxes, levies, remittances, etc. What are the effects of the downturn on government
budgets and debt (e.g. via fewer mining royalties, corporation tax, VAT, etc)? Are stresses
already occurring?
Prices: Changes in consumption and production prices, wages, salaries and interest rates.
Assets: Value of and access to and/or control of assets, including physical, natural, human,
social and financial assets.
Employment: Formal and informal employment, including self-employment and employment in
household enterprises. Other aspects include security, status, workloads and gender issues.
Investment: Changed capital flows, imports and export opportunities will affect fixed capital
investment. For example, lower copper prices will reduce investment in copper exploration.
Access to goods and services: Peoples access to public and private goods and services may
involve removal of barriers or improving the quality of goods and services available.
Taken together, these transmission mechanisms determine first-round effects on the poor and more
generally on the distribution of effects across incomes but also across space. To compute the final
effects, country studies need to report on how government policy has responded to changes in income.
Several Phase 1 countries reported actual employment and possible poverty implications. Table 10
shows the expected increase in the number of poor in 2009 as a result of the crisis, by taking changes
in GDP forecasts and multiplying these by number of poor people and poverty elasticity with respect to
growth. Countries may want to update this.
Table 11: Expected poverty effects of global financial crisis in 2009
Country

Mar 2009
Poverty
Household
forecast for
elasticity with
poverty
2009 IMF
respect to
count,
Spring 2008
growth
% (latest
(case studies
year, WDI)
GDP
projections
or literature)
for 2009
Bangladesh
-1.2
0.38
40
Bolivia
-1
0.2
65.20
Cambodia
-2.2
1
35
Ghana
-3.4
1
28.50
Kenya
-0.4
0.74
52
Uganda
-1
2
37.70
Zambia
-2.4
0.2
68
Source: Phase 1 country case studies and te Velde et al (2009)

Population,
millions in
2007 (WDI)

158.6
9.5
14.4
23.5
37.5
30.9
11.9

Poverty increase,
000s:
change in growth
rates,*
elasticity,*
household
poverty count
289.3
12.4
110.9
227.7
57.7
233.0
38.8

Loss of employment 2 is one feature of the global financial crisis in the case studies. Loss of
employment and income has important poverty and human development implications. The Phase 1
case studies suggested the following:
A recorded 15,000 construction workers in Cambodia were laid off in mid-2008. This has led to
some de-urbanisation. The garment industry has been hardest hit, with approximately 51,000
people laid off (many of these women) in the six months between September 2008 and March
2009.
2 It is possible to use the same framework as for poverty. The International Labour Organization (ILO) has estimated
employment elasticities with respect to growth and we have taken the average over 1991-2003: Nigeria, 1.11; Ghana, 0.77;
Zambia, 0.20; Cambodia, 0.37; Benin, 0.87; Bangladesh, 0.31; Bolivia, 1.26; Uganda, 0.34; Kenya, 1.96; Indonesia, 0.24.
14

In Zambia, 8100 workers in the mining sector lost their jobs in 2008.
A simulation exercise in Bolivia predicted rising unemployment as a result of lower remittances
(3.0%) and mining exports (1.6%).
Following an ILO methodology, incremental job opportunities in Bangladesh may be squeezed
by 500,000 compared with the expected level, if GDP growth slows down to the World Banks
projected level of 4.8%.
In Kenya, the labour-intensive horticultural industry, which employs an estimated 3 million
people, had to cut around 1200 jobs this year and suffered a 35% drop in exports of flowers.

The country case studies may want to update information on employment.

It was more difficult to report on price and asset effects of the crisis is this now possible?

15

4. Policy responses: A critical review

The crisis has affected countries since the end of 2008, and most countries have now seen the effects.
There were fears initially that countries would not respond, but it should now be possible to review
policy responses critically.

4.1

Macroeconomic policies to manage the impact of the crisis

Macro policies include:

Monetary and banking policies;
Fiscal policies.
By March 2009, Phase 1 countries had already put in place macroeconomic and financial policies to
manage the crisis:
The Central Bank of Kenya recently lowered the cash reserve ratio from 6% to 5% and the
central bank rate from 9% to 8.25% in order to lower interest rates and enhance credit supply in
the economy.
In Bolivia, the 2009 national budget included a 20.6% increase in public investment and a 12%
rise in public servants wages and salaries, accounting for an increase in government
consumption.
Cambodia has produced an expansionary budget, but there is discussion on the contents.
Central Bank of Bangladesh reserves were safeguarded through withdrawal from risky
investments and transfer to reliable central bank accounts, and private sector financial
institutions were advised to protect their respective deposits.
Country case studies are expected to provide an update on whether monetary and fiscal policies have
responded further, and review critically whether this has been appropriate. There is a need to
distinguish crisis responses from changes in policy for other reasons, e.g. the food price crisis, which
left monetary policy tighter than would otherwise have been the case.

4.2

Social policies to respond to the impact of the crisis

While the crisis has not yet resulted in major social protection policy revisions, or a large-scale
expansion of social protection provision in most countries, a number of pre-existing programmes have
been extended and new programmes introduced, albeit on a modest scale:
Putting in place or using safety nets and cash transfers for households affected by the global
financial crisis (or other recent crises);
Putting in place safety nets for households affected by the global financial crisis (or other
recent crises);
Changing allocations for social sectors, such as education and health.
The major types of interventions selected to date are food subsidies and rationing (Indonesia and
Bangladesh); food distributions for vulnerable groups and school feeding programmes (Cambodia,
Indonesia, Bangladesh, Ghana, Kenya and Nigeria); in-kind transfers offering fertiliser (Kenya); cash
transfers (Ghana); education scholarships and subsidies (Cambodia and Ghana); and public works
programmes (Cambodia, Indonesia, Bangladesh).
It is crucial to review social responses critically: were they put in place in a timely and well-targeted
manner, without sacrificing growth in the longer term?

16

4.3

Economy-wide and sectoral structural policies for getting the country out of
the crisis

There are limits to what can be done through short-term economic and social management, which is
unlikely to deal with structural challenges. Growth-enhancing policies are one way to get a country out
of the crisis. For example, a fiscal stimulus may bring spending and hence growth forward, but a
change in business conditions would lead to faster investment, which may raise growth now and in the
future.
It will be important to describe whether the crisis has led to changes in any policies that discriminate
among different types of sectors or groups in the economy, as well as looking at the changes in
aggregate fiscal stance. Structural policies could include:
Trade policies (tariffs, subsidies);
Tax policies (e.g. corporate taxes, investment incentives);
Competition policies;
Industrial policies (e.g. export processing zones, technology and R&D);
Business policies;
Investment climate and administrative procedures;
Human resource policies.
Such policies could have been implemented economy-wide (e.g. competition policy) or at sector level
(e.g. a new skills centre for garments). The aim of this sub-section is to monitor the policy responses
through structural growth approaches.

4.4

Institutional context and constraints of policymaking

Some countries have better statebusiness relations and institutional setups to respond to crisis. This
financial crisis has led to the setup of crisis task forces in some countries.
What is the composition of the task force?
What has it advocated?
What have been the effects?
Apart from the institutional setting, it may also be useful to discuss the economic constraints to
policymaking, e.g.
Exchange rate developments condition monetary policy responses.
Government debt conditions fiscal responses.

4.5

Multilateral and bilateral donor responses in-country

While Phase 1 results indicated no major pullout of aid, it is likely that there is some impact still to
come. More positively, the international system (e.g. IMF, regional development banks RDBs) has
responded quickly. It is possible to distinguish between bilateral aid responses and responses in
relation to the international system, including the international financial institutions (IFIs). One can
consider a number of issues when examining donor responses, such as scope of donor responses,
including changes to aid volumes, new aid instruments and reprogramming of aid programmes.
Multilaterals have been very active in responding to the crisis, but countries should respond on what
has happened in their country. The IMF has increased its lending to low-income countries from $1.2
billion in 2008 to an expected $4 billion this and next year. Conditionalities have tended to decrease
and the content of conditionalities has changed. However, responses (e.g. special drawing right SDR
allocations) may have varied across countries. The World Banks International Development

17

Association (IDA) and International Finance Corporation (IFC) have also increased their activity, at
differing speeds.
There are four ways to describe bilateral responses:
Aid volumes: Some countries have cut their aid spending, e.g. Italy by 56% and Ireland by
24%, leading to cuts in their ODA/GNI ratios. Other countries are sticking by their (high)
ODA/GNI levels (Denmark, the Netherlands), which implies cuts to planned ODA, e.g. by 600
million (or 11%) in the case of the Netherlands. There are also countries where aid is increasing,
such as the UK, with both volumes and ODA/GNI ratios increasing.
Aid programming: Some countries are changing the allocation of aid across countries owing to
the crisis (e.g. the Netherlands). It is not clear whether countries are changing their withincountry programmes. Others report that they have not made major changes (Germany). Some
are increasing their relative spending through multilateral channels (the Netherlands and
probably Norway).
New initiatives: Bilateral support to new, often multilateral, initiatives for the IMF or the World
Banks vulnerability fund (the Netherlands, Germany), and increased spending on social
protection (Australia) and infrastructure. There does not appear to be an overall bias and donor
interest in certain areas may prevail.
Countries should report what bilateral donors have done (including any changes planned for the
future).

18

5. Conclusions and looking ahead

5.1

The impact of the crisis: An update

This section concludes by asking what new insights have been gained from the new monitoring.

5.2

Looking ahead: How well is the country positioned to gain from a future
recovery and to grow sustainably?

The paper could conclude by looking into the future, asking whether the country is well positioned to
gain from a future recovery:
Is it competitive?
Has it introduced the right policies for future growth?
Has it the right institutional framework and capacity?
Will it be able to withstand future crises more effectively?

19

References
Cal, M., with S. DellErba, (2009) The Global Financial Crisis and Remittances: What Past Evidence
Suggests. Working Paper 303. London, UK: ODI.
Docquier, F. and H. Rapoport (2004) Skilled Migration: The Perspective of Developing Countries. Policy
Research Working Paper 3382. Washington, DC: World Bank.
International Monetary Fund (2009) The Implications of the Global Financial Crisis for Low-income
Countries. Washington, DC: IMF.
te Velde, D.W. (2009) The Global Financial Crisis and Developing Countries: An Update of the
Monitoring Work. London, UK: ODI.
te Velde, D.W. with C. Ackah, O. Ajakaiye, E. Aryeetey, D. Bhattacharya, M. Cal, T. Fakiyesi, A.G. Fulbert,
H. Jalilian, L.C. Jemio, J. Keane, J. Kennan, I. Massa, A. McCord, M. Meyn, M. Ndulo, M. Rahman, I.
Setiati, H. Soesastro, S. Ssewanyana, M. Vandemoortele and others (2009) The Global Financial
Crisis and Developing Countries: Synthesis of the Findings of 10 Country Case Studies. Working
Paper 306. London, UK: ODI.

20

Annex 1: Commodity prices

Figure A1: Price developments of crude oil (US$/bbl)
140
120
100
80
60
40
20

Source: World Bank Pink Sheets. Periods change from years, to quarters and months.

Figure A2: Price developments of aluminium (US$/mt)

3,500
3,000
2,500
2,000
1,500
1,000

Source: World Bank Pink Sheets. Periods change from years, to quarters and months.

Figure A3: Price developments of gold (US$/toz)

1,000
950
900
850
800
750
700
650
600
550

Source: World Bank Pink Sheets. Periods change from years, to quarters and months.

Figure A4: Price developments of copper (US$/mt)

8,000
7,000
6,000
5,000
4,000
3,000

Source: World Bank Pink Sheets. Periods change from years, to quarters and months.
21

Overseas Development Institute

111 Westminster Bridge Road
London SE1 7JD
UK
Tel: +44 (0)20 7922 0300
Fax: +44 (0)20 7922 0399
Email: publications@odi.org.uk
Website: www.odi.org.uk