## _wp04197

## Source details

**Canonical URL:** [_wp04197](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04197.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04197.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04197.pdf.json)

---

### Introduction
- Research divides into two strands: (1) immediate response to financial crises (stem crises, prevent contagion, regain market access) and (2) longer-term macroeconomic implications of IMF programs.
- Paper examines relationship between IMF programs and private international capital flows for 29 emerging market economies over 1980—2002, covering 119 programs (unit of analysis: program-country pair).
- Key methodological caution: the counterfactual—what would have happened without an IMF program—is central; selection bias and mean reversion must be addressed by comparing program and nonprogram countries with similar initial conditions.

### Channels through which the IMF can influence capital flows
- Good housekeeping “seal of approval”
  - Requires an informational advantage and credible, transparent communication.
  - Seal worked historically (gold standard) when credible commitment and institutions backed adherence.
  - Contemporary concerns: IMF informational advantage may have declined; possible political influence; some reasons IMF may retain an edge (multiple IMF economists covering a country; capacity to verify data and judge authorities’ intentions).
- Delegated monitor
  - Addresses asymmetric information and dual/common agency problems.
  - IMF programs impose conditionality and tranche-based lending to constrain opportunistic behavior and coordinate lenders.
  - Delegated monitoring likely most effective in an “intermediate range” of fundamentals (Mody and Saravia (2003) argument).
- Catalytic lending
  - IMF liquidity provision can lower default probability and induce private lenders to roll over loans, increasing gross capital flows.
  - Works if IMF can distinguish illiquidity from insolvency and choose appropriate lending amount.
  - Theoretical model predictions (Morris and Shin (2003); Corsetti, Guimarães, and Roubini (2003)):
    - Very bad fundamentals: IMF programs have no effect.
    - Bad fundamentals: IMF programs can induce adjustment, catalyze private rollovers, and improve fundamentals.
    - Good fundamentals: IMF programs may reduce country adjustment effort and have no catalytic effect.
    - Very good fundamentals: IMF programs have no effect.

### Aggregate trends before and after program adoption
- Sample and metrics
  - Sample: 29 emerging market economies (IMF International Capital Markets Department definition) over 1980–2002.
  - Performance metrics: real GDP growth, inflation, current account balance (percent of GDP), and capital flows (percent of GDP). Capital flows measured as aggregate gross bond, equity, and loan flows (Dealogic); other data from IMF sources.
- Balanced-sample approach: compare 12 months before program start (“Year Before IMF Program”) to first 12 months (“First Year”), second 12 months (“Second Year”), and third 12 months (“Third Year”) after program start.
- Main descriptive findings (summary)
  - Macroeconomic aggregates and capital flows improve following IMF program adoption, often with a lag.
  - Real GDP growth and capital flows exhibit a “dip and recovery” pattern:
    - First year: performance deteriorates somewhat (dip).
    - Second year: performance improves significantly (recovery).
    - Third year: weaker improvement or slight fallback but generally above pre-program levels.
  - Inflation: fails to improve in first year, large reduction in second and third years.
  - Current account: high initial deficit, reduced in first year and turns into a surplus in second year, then returns to a more sustainable deficit in third year.
- Program duration comparison (sample sizes: 90 SBA programs; 29 EFF programs)
  - EFFs associated with better performance than SBAs, including higher capital flows (with similar dip in first year).
- Selected exact aggregates from Table 2 (In percent)
  - EFF Programs (A)
    - Current account/GDP: –3.8, –2.7, 3.2, –3.2 (Year Before IMF Program, First Year, Second Year, Third Year)
    - Gross capital flows/GDP: 1.1, 1.0, 1.7, 1.5
    - Inflation: 23.5, 18.7, 13.5, 11.0
    - Real GDP growth: 1.4, 2.6, 5.2, 4.3
  - SBA Programs (B)
    - Current account/GDP: –3.7, –1.5, 1.7, –0.5
    - Gross capital flows/GDP: 2.1, 2.1, 2.2, 2.1
    - Inflation: 21.8, 25.4, 18.0, 15.3
    - Real GDP growth: 3.0, 1.8, 3.9, 3.9

### Performance conditional on initial fundamentals and lending levels
- Four external-fundamental indicators define initial states (Table 3 definitions preserved exactly, In percent unless otherwise noted):
  - Current Account to GDP:
    - 1 (Very bad) Deficit larger than 6
    - 2 (Bad) Deficit between 3 and 6
    - 3 (Good) Deficit between 0 and 3
    - 4 (Very good) Surplus
  - Reserves to Imports:
    - 1 (Very bad) Between 0 and 1.25
    - 2 (Bad) Between 1.25 and 3
    - 3 (Good) Between 3 and 7
    - 4 (Very good) More than 7
  - Short-term Debt to Reserves:
    - 1 (Very bad) Larger than 4
    - 2 (Bad) Between 2 and 4
    - 3 (Good) Between 1 and 2
    - 4 (Very good) Between 0 and 1
  - External Debt to GDP:
    - 1 (Very bad) Larger than 0.6
    - 2 (Bad) Between 0.4 and 0.6
    - 3 (Good) Between 0.2 and 0.4
    - 4 (Very good) Between 0 and 0.2
- Conditional capital flow patterns (selected figures from Table 4, In percent of GDP)
  - Conditional on Reserves/Imports (No. of Observations shown per state)
    - Very bad: 1.5, 1.2, 0.5, 0.5, No. of Observations 8
    - Bad: 1.5, 1.2, 2.1, 2.2, No. of Observations 16
    - Good: 2.2, 2.4, 2.5, 2.0, No. of Observations 11
    - Very good: 2.3, 2.4, 2.8, 2.8, No. of Observations 6
  - Conditional on Short-Term Debt/Reserves
    - Very bad: 1.6, 1.9, 1.6, 1.5
    - Bad: 1.6, 0.8, 1.8, 1.9
    - Good: 2.0, 2.6, 2.3, 1.9
    - Very good: 2.2, 2.2, 2.7, 2.6
  - Conditional on Current Account Deficit/GDP
    - Very bad: 2.4, 2.3, 1.8, 2.2
    - Bad: 2.1, 1.9, 2.7, 2.2
    - Good: 1.0, 1.2, 1.1, 1.1
    - Very good: 0.7, 1.2, 2.3, 1.6
  - Conditional on External Debt/GDP
    - Very bad: 1.5, 1.7, 1.6, 1.8
    - Bad: 1.4, 1.7, 1.9, 2.1
    - Good: 1.9, 1.7, 2.7, 1.9
    - Very good: 0.0, 1.1, 0.0, 0.5
- Summary of conditional findings
  - Very bad initial fundamentals:
    - IMF programs typically unable to reverse slide in capital flows; flows either fail to increase or continue to fall.
    - Example: reserves-to-imports in very bad state — three years after program start, capital flows-to-GDP ratio falls to one-third of initial level (Table 4 shows decline from 1.5 to 0.5).
    - For short-term debt-to-reserves in very bad state, capital flows may rise in first year but fall thereafter.
    - Very bad current account deficit cases: capital flows were substantial pre-program but decline up to two years post-program before rising.
    - Very bad external debt-to-GDP: slight indication of capital flow increases post-program (small sample).
  - Bad initial fundamentals:
    - Clear dip and recovery pattern for capital flows across most fundamentals (except external debt where flows increase steadily).
    - Second and third years show recovery above pre-program levels.
    - Liquidity indicators (reserves-to-imports; short-term debt-to-reserves) most clearly show dip-and-recovery.
  - Good and very good initial fundamentals:
    - Mixed evidence. Good initial state: IMF programs make little difference to capital flows trend.
    - Very good initial state: some cases show capital flows improve after program start (likely reflecting temporary liquidity issues prompting programs).
- Relationship between IMF lending amount and capital flows (selected panels from Table 5, In percent of GDP)
  - Amount/Reserves (No. of Observations)
    - Low: 1.8, 2.6, 1.3, 2.4, No. of Observations 6
    - Medium: 2.1, 1.9, 2.4, 1.9, No. of Observations 22
    - High: 1.1, 0.9, 1.5, 1.6, No. of Observations 13
  - Amount/Short-Term Debt
    - Low: 1.5, 2.1, 0.9, 1.1, No. of Observations 8
    - Medium: 2.4, 2.2, 2.3, 2.1, No. of Observations 12
    - High: 1.8, 1.5, 2.9, 2.6, No. of Observations 15
  - Composite Amount Index
    - Low: 2.2, 2.6, 1.6, 2.1, No. of Observations 11
    - Medium: 1.8, 1.4, 2.1, 1.8, No. of Observations 27
    - High: 1.1, 1.2, 2.2, 1.7, No. of Observations 10
  - Evidence inconclusive: larger IMF lending is associated with increases in capital flows over three years, but initial capital flows were lower where IMF lending was larger — causality unclear.

### Transition probabilities: program vs nonprogram countries
- State labeling: S_T: very bad (1), bad (2), good (3), very good (4).
- Construction
  - For each program-country pair: S_T = average state during year before program start (months t-11 through t); S_T+2 = average state during second year after program start (months t+13 through t+24).
  - For nonprogram countries: use January of each year to define S_T and S_T+2; include observations with no IMF program during months t-11 through t+24.
  - Sample sizes: approximately 250 nonprogram observations vs about 100 program observations.
- Hypothesis testing
  - Null: Pr(S_T+2 | S_T) is the same with or without an IMF program (two-sided tests; significance at the 5 percent level).
  - Appendix provides calculation of p-values: p1 denotes the transition probability with IMF program; p0 without; D = p1 − p0; σ_D and SE_D formulas given; z = (p̂1 − p̂0) / SE_D; p-value computed from standard normal.
- Selected exact two-year transition probabilities and p-values (Table 6, In percent)
  - Panel A. Current Account Balance to GDP (S_T = 1 Very bad)
    - Program S_T+2: 1: 0.23; 2: 0.15; 3: 0.38; 4: 0.23
    - Nonprogram S_T+2: 1: 0.29; 2: 0.49; 3: 0.17; 4: 0.06
    - P-value: 0.35, 0.01, 0.05, 0.03
  - Panel B. Reserves to Imports (S_T = 1 Very bad)
    - Program: 0.42, 0.17, 0.42, 0.00
    - Nonprogram: 0.71, 0.24, 0.05, 0.00
    - P-value: 0.04, 0.30, 0.00, n/a
  - Panel C. Short-Term Debt to Reserves (S_T = 1 Very bad)
    - Program: 0.20, 0.50, 0.30, 0.00
    - Nonprogram: 0.63, 0.32, 0.05, 0.00
    - P-value: 0.01, 0.15, 0.03, n/a
  - Additional p-values in Table 6 include: 0.01, 0.05, 0.50, 0.04, 0.12, 0.44, 0.21, 0.07, 0.49, 0.24, 0.37 (as presented in the source).
- Main findings from transition matrices
  - IMF programs are generally associated with improvements in external fundamentals after two years, except when fundamentals are already in state 4 (very good).
  - For countries starting in very bad state (state 1), IMF programs significantly increase the probability of moving to a better state after two years for several indicators.
  - For countries starting in bad state (state 2), IMF programs significantly increase the probability of moving to a better state after two years—except for the short-term debt-to-reserves indicator.
  - For countries starting in good state (state 3), IMF programs significantly increase the likelihood of improving to state 4 after two years for all external fundamentals.

### Synthesis, conclusions, and research suggestions
- Synthesis of empirical patterns
  - Three-channel framework reaffirmed: seal of approval, delegated monitoring, catalytic lending — with delegated monitoring and catalytic lending more important given limits to IMF informational advantage.
  - Descriptive before-after evidence: macro aggregates and capital flows improve after program adoption, with first-year dip and second-year recovery; inflation declines by second and third years; current account improves initially then moves to a more sustainable deficit by year three.
  - Program duration matters: EFFs (longer-term) associated with better outcomes than SBAs (short-term) in the sample.
- Conditional effectiveness by initial fundamentals (Table 7 qualitative summary)
  - Very bad (state 1): Either continued decline after initiation of program or, at best, no improvement. Some improvement in external fundamentals, but related to domestic recessionary conditions.
  - Bad (state 2): Dip and recovery, as the decline in capital flows is stemmed and a rise initiated. Strong evidence for improvement in external fundamentals.
  - Good and very good (states 3 and 4): Either no change or evidence consistent with stabilization of flows following a temporary shock. Ambiguous.
- Caveats and limitations
  - Counterfactuals remain imperfect; unobserved confounders may exist.
  - Sample restricted to emerging market economies, 1980–2002; findings not generalizable to all IMF programs.
  - Programs treated broadly uniform except for SBA vs EFF; program design and execution heterogeneity not fully explored.
- Suggestions for future research
  - Expand variables characterizing initial conditions to include real GDP growth, inflation, and capital flows themselves.
  - Use econometric methods (e.g., Markov switching models) to derive thresholds for states rather than prerefining them.
  - Differentiate IMF programs by design (precautionary vs nonprecautionary) and execution (cancelled vs completed).
  - Model determinants of capital flows more fully, including demand-side factors and measures of gross financing need; distinguish “pull” vs “push” drivers.

*Source: _wp04197 (PDF chapter content provided).*

### References..............................................................................................................

### _wp04197 - References

### Appendix
- Calculation of p-Values for Differences in Transition Probabilities ................................................................................................................. 26

### Tables
- 1. Summary of Predictions by Morris and Shin (2003) and Corsetti, Guimaraes, and Roubini (2003) ................................................................................................... 27
- 2. Average Performance of Emerging-Market Economies With EEF and SBA Programs ................................................................................................................... 28
- 3. Definitions of States in Terms of Fundamentals........................................................................................ 29
- 4. Average Capital Flows to Emerging-Market Economies Conditional on External Fundamentals.............................................................................................. 30
- 5. Average Capital Flows to Emerging-Market Economies Conditional on IMF Lending Amount ............................................................................................... 31
- 6. Two-Year Transition Probabilities for Emerging-Market Economies With and Without IMF Programs ............................................................................................. 32
- 7. Summary of Main Results ......................................................................................... 33

*Source: _wp04197 - References (page listings and appendix title as provided).*

### 1. Average Performance of Emerging-Market Economies with IMF Programs ........... 34

### _wp04197 - 1. Average Performance of Emerging-Market Economies with IMF Programs ........... 34

### Introduction
- Research on the IMF divides into two strands: (1) immediate response to financial crises (stem crises, prevent contagion, regain market access) and (2) longer-term macroeconomic implications of IMF programs.
- Three research challenges highlighted:
  - Bring the two strands closer together (do IMF programs serve as emergency finance only, or also improve fundamentals and long-term market access?).
  - Account for differences in countries’ initial economic conditions and potential non-linear responses.
  - Identify channels/mechanisms through which the IMF influences outcomes to inform policy design and financial architecture.
- This paper examines the relationship between IMF programs and private international capital flows for 29 emerging market economies over 1980—2002, covering 119 programs (unit of analysis: program-country pair).
- Key methodological caution: the counterfactual—what would have happened without an IMF program—is central; selection bias and mean reversion must be addressed by comparing program and nonprogram countries with similar initial conditions.

### The role of the IMF in facilitating capital flows — channels of influence
- The IMF can influence capital flows through three channels:
  1. Good housekeeping “seal of approval”
     - Requires an informational advantage and credible, transparent communication.
     - Seal worked historically (gold standard) when credible commitment and institutions backed adherence.
     - Contemporary concerns: IMF informational advantage may have declined (greater transparency by IMF and markets; no clear superiority of IMF forecasts; possible political influence).
     - Some reasons IMF may retain an edge: multiple IMF economists covering a country; capacity to verify data and judge authorities’ intentions.
  2. Delegated monitor
     - Addresses asymmetric information where governments may behave opportunistically (dual agency) or where multiple lenders lead to overborrowing (common agency).
     - IMF programs impose conditionality and tranche-based lending to constrain opportunistic behavior and coordinate lenders.
     - Delegated monitoring likely most effective in an “intermediate range” of fundamentals (Mody and Saravia (2003) argument).
  3. Catalytic lending
     - IMF liquidity provision can lower default probability and induce private lenders to roll over loans, increasing gross capital flows.
     - Works if the IMF can distinguish illiquidity from insolvency and choose an appropriate lending amount.
     - Theoretical model predictions (Morris and Shin (2003); Corsetti, Guimarães, and Roubini (2003)):
       - Very bad fundamentals: IMF programs have no effect.
       - Bad fundamentals: IMF programs can induce adjustment, catalyze private rollovers, and improve fundamentals.
       - Good fundamentals: IMF programs may reduce country adjustment effort and have no catalytic effect.
       - Very good fundamentals: IMF programs have no effect.

### Aggregate trends before and after program adoption
- Sample: 29 emerging market economies (IMF International Capital Markets Department definition) over 1980–2002.
- Performance metrics: real GDP growth, inflation, current account balance (percent of GDP), and capital flows (percent of GDP). Capital flows measured as aggregate gross bond, equity, and loan flows (Dealogic); other data from IMF sources.
- Balanced-sample approach: compare 12 months before program start (“Year Before IMF Program”) to first 12 months (“First Year”), second 12 months (“Second Year”), and third 12 months (“Third Year”) after program start; years defined precisely around program start month.
- Main descriptive findings:
  - Macroeconomic aggregates and capital flows improve following IMF program adoption, often with a lag.
  - Real GDP growth and capital flows exhibit a “dip and recovery” pattern:
    - First year: performance deteriorates somewhat (dip).
    - Second year: performance improves significantly (recovery).
    - Third year: weaker improvement or slight fallback but generally above pre-program levels.
  - Inflation: fails to improve in first year, large reduction in second and third years.
  - Current account: high initial deficit, reduced in first year and turns into a surplus in second year, then returns to a more sustainable deficit in third year.
- Program duration comparison:
  - Stand-By Arrangements (SBAs): typically 12–18 months; repayment expected within 2¼ to 4 years.
  - Extended Fund Facility (EFF): typically about 3 years; repayment expected within 4½ to 7 years.
  - Evidence suggests EFFs are associated with better performance than SBAs (sample sizes: 90 SBA programs versus 29 EFF programs), including higher capital flows (with similar dip in first year). Explanations: longer commitment, stronger seal of approval, longer delegated monitoring, often more lending, or longer repayment horizons.

### Performance conditional on initial fundamentals and lending levels
- Four external-fundamental indicators used to define initial conditions (states: “very bad,” “bad,” “good,” “very good”):
  - Current account balance as percent of GDP.
  - Reserves-to-imports ratio.
  - Ratio of short-term debt to reserves.
  - Ratio of total external debt to GDP.
- Thresholds (illustrative, based on common literature/IMF practice; definitions summarized in Table 3 of source):
  - Very bad:
    - Current account deficit > 6 percent of GDP
    - Reserves < 1.25 months of imports
    - Short-term debt > four times reserves
    - Total external debt > 60 percent of GDP
- Conditional results on capital flows (Table 4 summary):
  - Very bad initial fundamentals:
    - IMF programs typically unable to reverse slide in capital flows; flows either fail to increase or continue to fall.
    - Example: reserves-to-imports in very bad state — three years after program start, capital flows-to-GDP ratio falls to one-third of initial level.
    - For short-term debt-to-reserves in very bad state, capital flows may rise in first year but fall thereafter.
    - Very high current account deficit cases: capital flows were substantial pre-program but decline up to two years post-program before rising.
    - Very bad external debt-to-GDP: slight indication of capital flow increases post-program (small sample; likely not statistically significant).
  - Bad initial fundamentals:
    - Clear dip and recovery pattern for capital flows across most fundamentals (except external debt where flows increase steadily).
    - Second and third years show recovery above pre-program levels — IMF programs appear effective in helping transition from vulnerability to improvement.
    - Liquidity indicators (reserves-to-imports; short-term debt-to-reserves) most clearly show dip-and-recovery.
  - Good and very good initial fundamentals:
    - Mixed evidence. Good initial state: IMF programs make little difference to capital flows trend (consistent with theoretical predictions).
    - Very good initial state: some cases show capital flows improve after program start (likely reflecting temporary liquidity issues prompting programs).
- Relationship between size of IMF lending and capital flows (Table 5):
  - Evidence inconclusive: larger IMF lending is associated with increases in capital flows over three years, but initial capital flows were lower where IMF lending was larger — unclear causality.

### Transition probabilities: program vs nonprogram countries
- Objective: approximate counterfactual by comparing transition probability matrices across the four states for program and nonprogram countries.
- State variable S_T: very bad (1), bad (2), good (3), very good (4).
- For each program-country pair:
  - S_T = average state during year before program start (months t-11 through t).
  - S_T+2 = average state during second year after program start (months t+13 through t+24).
- For nonprogram countries:
  - Use January of each year to define S_T and S_T+2, and only include observations where no IMF program was in place during months t-11 through t+24.
  - Approximately 250 nonprogram observations vs about 100 program observations.
- Null hypothesis: Pr(S_T+2 | S_T) is the same with or without an IMF program (two-sided tests; significance at the 5 percent level).
- Main results (Table 6 summary):
  - IMF programs are generally associated with improvements in external fundamentals after two years, except when fundamentals are already in state 4 (very good).
  - Even for countries starting in very bad state (state 1), IMF programs significantly increase the probability of moving to a better state after two years.
  - For countries starting in bad state (state 2), IMF programs significantly increase the probability of moving to a better state after two years — except for the short-term debt-to-reserves indicator.
  - For countries starting in good state (state 3), IMF programs significantly increase the likelihood of improving to state 4 after two years for all external fundamentals.

### Synthesis and conclusions
- Three-channel framework reaffirmed: seal of approval, delegated monitoring, catalytic lending — with delegated monitoring and catalytic lending more important in practice given limits to IMF informational advantage.
- Descriptive before-after evidence: macro aggregates and capital flows improve after program adoption, with first-year dip and second-year recovery; inflation declines by second and third years; current account improves initially then moves to a more sustainable deficit by year three.
- Program duration matters: EFFs (longer-term) associated with better outcomes than SBAs (short-term) in the sample.
- Conditional effectiveness by initial fundamentals:
  - Very bad fundamentals: IMF programs at best stem deterioration in fundamentals but typically do not revive capital flows; improvements may reflect import compression and may not be sustainable.
  - Bad (vulnerable) fundamentals: IMF programs most clearly associated with improved capital flows and improved external fundamentals over two- to three-year horizons — strongest evidence of IMF benefit.
  - Good/very good fundamentals: IMF programs may help stabilize capital flows after temporary shocks; effects on fundamentals ambiguous; moral hazard concerns not strongly supported by evidence here.
- Caveats and limitations:
  - Counterfactuals remain imperfect; unobserved confounders may exist.
  - Sample restricted to emerging market economies, 1980–2002; findings not generalizable to all IMF programs.
  - Programs treated broadly uniform except for SBA vs EFF; program design and execution heterogeneity not fully explored.
- Suggestions for future research:
  - Expand variables characterizing initial conditions to include real GDP growth, inflation, and capital flows themselves.
  - Use econometric methods (e.g., Markov switching models) to derive thresholds for states rather than prerefining them.
  - Differentiate IMF programs by design (precautionary vs nonprecautionary) and execution (cancelled vs completed).
  - Model determinants of capital flows more fully, including demand-side factors and measures of gross financing need; distinguish “pull” vs “push” drivers.

*Source: _wp04197 - 1. Average Performance of Emerging-Market Economies with IMF Programs ........... 34 (PDF chapter).*

### REFERENCES

### _wp04197 - REFERENCES

### References (selected entries as listed)
- Barro, Robert J., and Jong-Wha Lee, 2003, “IMF Programs: Who Is Chosen and What Are the Effects?,” Working Papers in Trade and Development, No. 2003/09 (Canberra, Australia: Australian National University). Available via the Internet: http://post.economics.harvard.edu/faculty/barro/papers/IMF_Programs.pdf
- Bird, Graham, and Dane Rowlands, 1997, “The Catalytic Effect of Lending by the International Financial Institutions,” World Economy, Vol. 20 (No. 7), pp. 967–91.
- Boorman, Jack, and Mark Allen, 2000, “A New Framework for Private Sector Involvement in Crisis Prevention and Crisis Management,” in Reforming the International Financial System: Crisis Prevention and Response, ed. by Jan Joost Teunissen (The Hague: Forum on Debt and Development), pp. 101–23.
- Bordo, Michael D., 2003, “Market Discipline and Financial Crisis Policy: A Historical Perspective, in Research in Financial Services: Private and Public Policy, Vol.15, ed. by George G. Kaufman. (Greenwich, Connecticut, JAI Press), pp.157–182.
- Bordo, Michael D., and Harold James, 2000, “The International Monetary Fund: Its Present Role in Historical Perspective,” NBER Working Paper No. 7724 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Bordo, Michael D. and Finn E. Kydland, 1995, “The Gold Standard as a Rule: An Essay in Exploration,” Explorations in Economic History, Vol. 32 (No. 4), pp. 423–64.
- Bordo, Michael D., and Finn E. Kydland, 1996, “The Gold Standard as a Commitment Mechanism,” in Modern Perspectives on the Gold Standard, ed. by Tamim Bayoumi, Barry Eichengreen, and Mark P. Taylor (Cambridge, New York, and Melbourne: Cambridge University Press), pp. 55–100.
- Bordo, Michael D., and Hugh Rockoff, 1996, “The Gold Standard as a ‘Good Housekeeping Seal of Approval,’” Journal of Economic History, Vol. 56 (No. 2), pp. 389–428.
- Bordo, Michael D., and Anna J. Schwartz, 2000, “Measuring Real Economic Effects of Bailouts: Historical Perspectives on How Countries in Financial Distress Have Fared With and Without Bailouts,” NBER Working Paper No. 7701 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Calvo, Guillermo A., 1998, “Capital Flows and Capital-Market Crises: The Simple Economics of Sudden Stops,” Journal of Applied Economics, Vol. 1 (No. 1), pp. 35–54.
- Corsetti, Giancarlo, Bernardo Guimarães, and Nouriel Roubini, 2003, “International Lending of Last Resort and Moral Hazard: A Model of IMF's Catalytic Finance,” NBER Working Paper No. 10125 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Cottarelli, Carlo, and Curzio Giannini, 2002, “Bedfellows, Hostages, or Perfect Strangers? Global Capital Markets and the Catalytic Effect of IMF Crisis Lending,” IMF Working Paper 02/193 (Washington: International Monetary Fund).
- Ferguson, Roger W., 2002, “Understanding Financial Consolidation,” Economic Policy Review, Federal Reserve Bank of New York, Vol. 8 (May), pp. 209–13.
- Ghosh, Atish R., and others, 2002, IMF-Supported Programs in Capital Account Crises, IMF Occasional Paper No. 210 (Washington: International Monetary Fund).
- Goldstein, M., and P. Montiel, 1986, “Evaluating Fund Stabilization Programs with Multicountry Data: Some Methodological Pitfalls,” Staff Papers, Vol. 33 (June), pp. 304-44 (Washington: International Monetary Fund).
- Hajro, Zlata, and Joseph P. Joyce, 2004, “A True Test: Do IMF Programs Hurt the Poor?” Wellesley College Department of Economics Working Paper No. 04-01 (Wellesley, Massachusetts: Wellesley College).
- Haque, Nadeem Ul, and Mohsin Khan, 1998, “Do IMF-Supported Programs Work? A Survey of the Cross-Country Empirical Evidence,” IMF Working Paper 98/169 (Washington: International Monetary Fund).
- Hutchison, M., 2001, “A Cure Worse Than the Disease? Currency Crises and the Output Costs of IMF-Supported Stabilization Programs,” NBER Working Paper No. 8305 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Ivanova, Anna, 2003, “IMF Conditionality and Implementation of IMF-Supported Programs” (Ph.D. dissertation; Madison, Wisconsin: University of Wisconsin-Madison).
- Khan, M., 1990, “The Macroeconomic Effects of Fund-Supported Adjustment Programs,” Staff Papers, Vol. 37 (June), pp. 195–231 (Washington: International Monetary Fund).
- Köhler, Horst, 2001, “Promoting Stability and Prosperity in a Globalized World,” remarks by IMF Managing Director, Council of the Americas, Washington, May 7. Available via the Internet: http://www.imf.org/external/np/speeches/2001/050701.htm
- Loungani, Prakash, 2001, “How Accurate Are Private Forecasts? Cross-Country Evidence from Consensus Forecasts of Output Growth, International Journal of Forecasting, Vol. 17 (July-September), pp. 419–32.
- Lerrick, Adam, and Allan H. Meltzer, 2003, “Blueprint for an International Lender of Last Resort,” Journal of Monetary Economics, Vol. 50 (January), pp. 289–303.
- Marchesi, Silvia, and Jonathan P. Thomas, 1999, “IMF Conditionality as a Screening Device,” Economic Journal, Vol. 109 (March), pp. C111–25.
- Masson, Paul R., and Michael Mussa, 1995, “Long-Term Tendencies in Budget Deficits and Debt,” IMF Working Paper 95/128 (Washington: International Monetary Fund).
- Meltzer, Allan H., 2002, “New International Financial Arrangements,” Monetary and Economic Studies, Vol. 20 (special edition, December 2002), pp. 11–22.
- Miller, Marcus, and Lei Zhang, 2000, “Sovereign Liquidity Crises: The Strategic Case for a Payments Standstill,” Economic Journal, Vol. 110 (January), pp. 335–62.
- Mody, Ashoka, and Diego Saravia, 2003, “Catalyzing Capital Flows: Do IMF-Supported Programs Work as Commitment Devices?” IMF Working Paper 03/100 (Washington: International Monetary Fund).
- Morris, Stephen Edward, and Hyun Song Shin, 2003, “Catalytic Finance: When Does It Work?” Cowles Foundation Discussion Paper No. 1400 (New Haven: Yale University). Available via the Internet: http://ssrn.com/abstract=383604
- Obstfeld, Maurice, and Alan M. Taylor, 2003, “Sovereign Risk, Credibility and the Gold Standard: 1870-1913 versus 1925-31,” Economic Journal, Vol. 113 (No. 487), pp. 241–75.
- Powell, Andrew, and Leandro Arozamena, 2003, “Liquidity Protection Versus Moral Hazard: The Role of the IMF,” Journal of International Money and Finance, Vol. 22 (special issue, December), pp. 1041–63. Available via the Internet: http://www.utdt.edu/~apowell/
- Przeworski, Adam, and James Raymond Vreeland, 2000, “The Effect of IMF Programs on Economic Growth,” Journal of Development Economics Vol. 62 (August), pp. 385–421.
- Reinhart, Carmen M., Kenneth Rogoff, and Miguel Savastano, 2003, “Debt Intolerance,” Brookings Papers on Economic Activity: 1 (Washington: Brookings Institution), pp. 1–74.
- Rodrik, Dani, 1996, “Why Is There Multilateral Lending?” in Annual World Bank Conference on Development Economics, 1995, ed by Michael Bruno and Boris Pleskovic, (Washington: The World Bank), pp 167–93.
- Tirole, Jean, 2002, Financial Crises, Liquidity, and the International Monetary System (Princeton and Oxford: Princeton University Press).

### Appendix: Calculation of p-Values for Differences in Transition Probabilities
- Notation:
  - p1 denotes the transition probability between two states for countries with an IMF program.
  - p0 denotes the transition probability between the same two states for countries without an IMF program.
  - D = p1 – p0 denotes the difference between the two transition probabilities.
  - n0 and n1 are the number of observations corresponding to each row in the transition probability matrix (e.g., for p11, this is the number of cases where S_T = 1).
- Standard deviation of D:
  - σ_D = sqrt( (p00(1−p00)/n0) + (p11(1−p11)/n1) ) — as given in the source (expressed in the source as an equation with terms pp and nn).
- Under the null hypothesis H0: p1 = p0 = p. Let p̂0, p̂1, and p̂ denote the estimates (sample proportions).
  - p̂ = (X0 + X1) / (n0 + n1), where X0 and X1 are the number of transitions without and with an IMF program, respectively.
- Estimated standard error of D under H0:
  - SE_D = sqrt( p̂(1−p̂) (1/n0 + 1/n1) ) — expressed in the source as ˆDp SEpp nn  =− +   (notation preserved in text).
- Test statistic and p-value:
  - z = (p̂1 − p̂0) / SE_D, approximately N(0,1) if n0 and n1 are large.
  - p-value for H0: p1 = p0 against Ha: p1 ≠ p0 is Pr(Z ≠ z), where Z is a standard normal random variable.

### Key Tables and Numerical Results (figures and selected table entries preserved exactly)
- Table 1. Summary of Predictions by Morris and Shin (2003) and Corsetti, Guimaraes, and Roubini (2003)
  - (Table structure reproduced in the source showing combinations of Effect of IMF lending on adjustment effort, Effect of IMF lending on capital flows, and Fundamentals with labels Very Bad, Bad, Good, Very Good.)

- Table 2. Average Performance of Emerging-Market Economies With EFF and SBA Programs (In percent)
  - A. Average Performance of Emerging Market Economies with EFF Programs
    - Fundamental / Year Before IMF Program / First Year / Second Year / Third Year
    - Current account/GDP –3.8 –2.7 3.2 –3.2
    - Gross capital flows/GDP 1.1 1.0 1.7 1.5
    - Inflation 23.5 18.7 13.5 11.0
    - Real GDP growth 1.4 2.6 5.2 4.3
  - B. Average Performance of Emerging Market Economies with SBA Programs
    - Current account/GDP –3.7 -1.5 1.7 –0.5
    - Gross capital flows/GDP 2.1 2.1 2.2 2.1
    - Inflation 21.8 25.4 18.0 15.3
    - Real GDP growth 3.0 1.8 3.9 3.9

- Table 3. Definitions of States in Terms of Fundamentals (In percent)
  - Current Account to GDP:
    - 1 (Very bad) Deficit larger than 6
    - 2 (Bad) Deficit between 3 and 6
    - 3 (Good) Deficit between 0 and 3
    - 4 (Very good) Surplus
  - Reserves to Imports:
    - 1 (Very bad) Between 0 and 1.25
    - 2 (Bad) Between 1.25 and 3
    - 3 (Good) Between 3 and 7
    - 4 (Very good) More than 7
  - Short-term Debt to Reserves:
    - 1 (Very bad) Larger than 4
    - 2 (Bad) Between 2 and 4
    - 3 (Good) Between 1 and 2
    - 4 (Very good) Between 0 and 1
  - External Debt to GDP:
    - 1 (Very bad) Larger than 0.6
    - 2 (Bad) Between 0.4 and 0.6
    - 3 (Good) Between 0.2 and 0.4
    - 4 (Very good) Between 0 and 0.2

- Table 4. Average Capital Flows to Emerging-Market Economies Conditional on External Fundamentals (In percent of GDP) — selected panels
  - A. Conditional on Reserves/Imports (No. of Observations shown per state)
    - Very bad: Year Before IMF Program 1.5, First Year 1.2, Second Year 0.5, Third Year 0.5, No. of Observations 8
    - Bad: 1.5, 1.2, 2.1, 2.2, No. of Observations 16
    - Good: 2.2, 2.4, 2.5, 2.0, No. of Observations 11
    - Very good: 2.3, 2.4, 2.8, 2.8, No. of Observations 6
  - B. Conditional on Short-Term Debt/Reserves
    - Very bad: 1.6, 1.9, 1.6, 1.5
    - Bad: 1.6, 0.8, 1.8, 1.9
    - Good: 2.0, 2.6, 2.3, 1.9
    - Very good: 2.2, 2.2, 2.7, 2.6
  - C. Conditional on Current Account Deficit/GDP
    - Very bad: 2.4, 2.3, 1.8, 2.2
    - Bad: 2.1, 1.9, 2.7, 2.2
    - Good: 1.0, 1.2, 1.1, 1.1
    - Very good: 0.7, 1.2, 2.3, 1.6
  - D. Conditional on External Debt/GDP
    - Very bad: 1.5, 1.7, 1.6, 1.8
    - Bad: 1.4, 1.7, 1.9, 2.1
    - Good: 1.9, 1.7, 2.7, 1.9
    - Very good: 0.0, 1.1, 0.0, 0.5

- Table 5. Average Capital Flows to Emerging-Market Economies Conditional on IMF Lending Amount (In percent of GDP) — selected panels
  - A. Amount/Reserves
    - Low: 1.8, 2.6, 1.3, 2.4, No. of Observations 6
    - Medium: 2.1, 1.9, 2.4, 1.9, No. of Observations 22
    - High: 1.1, 0.9, 1.5, 1.6, No. of Observations 13
  - B. Amount/Short-Term Debt
    - Low: 1.5, 2.1, 0.9, 1.1, No. of Observations 8
    - Medium: 2.4, 2.2, 2.3, 2.1, No. of Observations 12
    - High: 1.8, 1.5, 2.9, 2.6, No. of Observations 15
  - C. Amount/Imports
    - Low: 2.2, 1.9, 1.7, 2.2, No. of Observations 11
    - Medium: 2.2, 2.4, 2.3, 2.0, No. of Observations 12
    - High: 1.1, 1.1, 1.9, 1.5, No. of Observations 18
  - D. Composite Amount Index
    - Low: 2.2, 2.6, 1.6, 2.1, No. of Observations 11
    - Medium: 1.8, 1.4, 2.1, 1.8, No. of Observations 27
    - High: 1.1, 1.2, 2.2, 1.7, No. of Observations 10

- Table 6. Two-Year Transition Probabilities for Emerging-Market Economies With and Without IMF Programs (In percent)
  - Panel A. Current Account Balance to GDP (S_T = 1 Very bad)
    - Program: S_T+2 = 1: 0.23; =2: 0.15; =3: 0.38; =4: 0.23
    - Nonprogram: S_T+2 = 1: 0.29; =2: 0.49; =3: 0.17; =4: 0.06
    - P-value: 0.35, 0.01, 0.05, 0.03
  - Panel B. Reserves to Imports (S_T = 1 Very bad)
    - Program: 0.42, 0.17, 0.42, 0.00
    - Nonprogram: 0.71, 0.24, 0.05, 0.00
    - P-value: 0.04, 0.30, 0.00, n/a
  - Panel C. Short-Term Debt to Reserves (S_T = 1 Very bad)
    - Program: 0.20, 0.50, 0.30, 0.00
    - Nonprogram: 0.63, 0.32, 0.05, 0.00
    - P-value: 0.01, 0.15, 0.03, n/a
  - (Additional rows and panels in Table 6 include probabilities and P-values for S_T = 2, 3, 4 across Program and Nonprogram groupings; p-values shown include 0.01, 0.05, 0.50, 0.04, 0.12, 0.44, 0.21, 0.07, 0.49, 0.24, 0.37 as presented in the source.)

- Table 7. Summary of Main Results (qualitative)
  - Initial Fundamentals / Capital Flow Trends / Change in Fundamentals
    - Very bad (state 1): Either continued decline after initiation of program or, at best, no improvement. Some improvement in external fundamentals, but related to domestic recessionary conditions.
    - Bad (state 2): Dip and recovery, as the decline in capital flows is stemmed and a rise initiated. Strong evidence for improvement in external fundamentals.
    - Good and very good (states 3 and 4): Either no change or evidence consistent with stabilization of flows following a temporary shock. Ambiguous.

### Figures (as labeled)
- Figure 1. Average Performance of Emerging-Market Economies with IMF Programs (panels and axes shown in the source):
  - A: Real GDP Growth — axis ticks: 0 1 2 3, vertical percentages 1% 2% 3% 4% 5% 6%
  - B: Inflation — axis ticks: 0 1 2 3, vertical percentages 6% 10% 14% 18% 22% 26% 30%
  - C: Current Account / GDP — axis ticks: 0 1 2 3, vertical percentages -4% -3% -2% -1% 0% 1% 2% 3% 4%
  - D: Gross Capital Flows/GDP — axis ticks: 0 1 2 3, vertical percentages 0.8% 1.2% 1.6% 2.0% 2.4%

*Source: _wp04197 - REFERENCES (source PDF content provided).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04197.pdf_