## _wp04104 ## Source details **Canonical URL:** [_wp04104](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04104.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04104.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04104.pdf.json) --- ### Introduction and main empirical findings - Aggregate investment exhibits large cyclical movements that can arise from firms' behavior under uncertainty rather than only from credit-market propagation mechanisms. - Using Italian manufacturing firms, evidence is found of: - Heterogeneity across firms in investment dynamics. - Slower adjustment of investment in response to demand shocks at higher levels of uncertainty. - An additional source of nonlinearity: a convex response of investment to demand shocks. - Policy implication: - A given demand stimulus will tend to have weaker short-run effects on investment when firms operate in a more uncertain environment. ### Theoretical background and predictions - Models of irreversible investment under uncertainty (real options) predict: - A wider inaction region under higher uncertainty, implying reduced short-run responsiveness of investment to demand shocks. - Threshold (trigger) policies: investment only when the gap between current and optimal capital is sufficiently large. - In multi-plant or multi-line settings with supermodularity, a convex response of firm-level investment to positive demand shocks is predicted. - Long-run effect of uncertainty on average capital stock is theoretically ambiguous (user-cost vs. hangover effects). ### Econometric specification and hypotheses tested - Baseline: error-correction model (ECM) derived from an ADL representation; dynamic panel framework. - Augmented ECM (equation (5) in the source) includes: - Linear and quadratic real sales growth terms (y_it and (y_it)^2) to test for convexity/nonlinearity. - Interaction term (_it y_it) to test whether short-run response to demand shocks depends on measured uncertainty. - Long-run uncertainty term (^2_{i;t-1}) to test for persistent effects on capital stock level. - Control for cash flow (C_it / K_{i;t-1}) to avoid confounding liquidity effects. - Error structure: firm fixed effects (_i), time dummies (_t), and GMM-style instrumentation to address endogeneity. ### Data, sample, and stylized facts - Data sources: - Survey of Investment in Manufacturing (SIM) from Bank of Italy (annual, representative sample of manufacturing firms with more than 50 employees), electronic format since 1984. - Company Accounts Data Service (CADS) from Centrale dei Bilanci. - Sample construction and coverage: - Merged SIM and CADS; loss of approximately one-third of initial 14,873 observations. - Final sample: Observations 4,192; Firms 564. - Median firm size in SIM-CADS sample: 266 employees (Firm size (employees median)285266 in Table A1). - Period coverage: SIM data available 1984–1998; tables and analyses often use 1991–1998 or 1992–1995 for specific breakdowns. - Stylized facts on zero-investment episodes (selected entries verbatim): - Table 1 (period 1991–1998, 2,682 observations), Employees 50ñ99: Buildings Investment 53.30, Machinery Investment 4.95, Transport Investment 41.75, Total Investment 54.17, Total Disinvestment 45.52, Investment and Disinvestment 3.30. - Table 2 (period 1992–1995, 1,689 observations), Plants =1: Building Investment 38.96, Machinery Investment 2.19, Transport Investment 23.75, Total Investment 51.46, Total Disinvestment 35.57, Investment and Disinvestment 1.20. ### Measuring uncertainty - Firm-level uncertainty measure constructed from managers' one-year-ahead forecasts of their own investment: - Forecast error (f e)_it = I_it / K_{i;t-1} − E_{i;t-1}[I_it / K_{i;t-1}] (equation (6)). - Squared forecast error (f e)^2_it (equation (7)). - Rolling variance (time-varying) used as main measure: ^2_it = (1/t) Σ_{s=1}^t (f e)^2_is (equation (9)), estimated using at least 4 observations (t ≥ 4). - Descriptive relationships (selected statistics from tables): - Sample mean I_it / K_{i;t-1} = 0.0962; Low Uncertainty mean = 0.1174; High Uncertainty mean = 0.0749. - Sample mean E_{i;t-1}[I_it / K_{i;t-1}] = 0.0937. - Firms with higher measured uncertainty tend to invest less (and expect to invest less) in proportion to capital stock. - Table 4 sectoral and size breakdown (examples): Electrical goods Low Unc. 42:86, High Unc. 57:14; Machinery Low Unc. 46:32, High Unc. 53:68. Size: 50ñ99 employees Low Unc. 46:21, High Unc. 53:79; 100 or more employees Low Unc. 50:00, High Unc. 50:00. ### Estimation method and diagnostics - Dynamic panel estimation using GMM (system GMM: Arellano-Bover/Blundell-Bond), with both one-step and two-step estimators; Windmeijer (2000) finite sample correction applied to two-step standard errors. - Instruments: lagged levels and lagged differences of endogenous variables; current sales treated as predetermined in preferred specification. - Diagnostic tests reported (as tabulated): - Sargan (p): 0:34, 0:37, 0:46, 0:44, 0:34. - LM2 (p): one-step: 0:71, 0:83, 0:77, 0:76, 0:86; two-step: 0:74, 0:78, 0:64, 0:64, 0:71. ### Key empirical results (one-step and two-step GMM) - Sample size and firms: Observations 4;192; Firms 564. - One-step results (selected coefficients from Table 5): - Column (1) basic specification: - y_it: 0:1163 (std. err. 0:0588) - y_{i;t-1}: 0:1456 (0:0239) - (k−y)_{i;t-2}: −0:1354 (0:0273) - Column (2) includes quadratic term: - y^2_it: −0:3376 (0:1281) (table formatting shows coefficient labeled with leading minus sign for display; text notes later signs reported as positive). - Column (3) includes interaction ^2_it y_it: - ^2_it y_it: −0:9209 (0:3862) (indicating weaker response at higher uncertainty). - Column (4) adds long-run uncertainty term: - ^2_{i;t-1}: −0:0199 (0:0545). - Two-step results (selected coefficients from Table 6): - Column (1) basic specification: - y_it: 0:1030 (0:0255) - y_{i;t-1}: 0:1228 (0:0219) - (k−y)_{i;t-2}: −0:1314 (0:0228) - Column (2) quadratic term: - y^2_it: −0:2258 (0:1021) - Column (3) interaction: - ^2_it y_it: −0:8247 (0:3439) - Column (4) long-run uncertainty term: - ^2_{i;t-1}: −0:0115 (0:0477) - Interpretation emphasized: - The interaction coefficient (one-step point estimate 0:92; two-step 0:8247) is large in absolute value and statistically different from zero, consistent with the theoretical prediction of weaker short-run responses under higher uncertainty. - Evidence of convexity (nonlinear response) is reported (authors note one-step point estimate 0:34 on quadratic term in text), supporting a convex response of investment to demand shocks. - Cash flow was not statistically significant in any specification. - Constant returns to scale hypothesis was not rejected and imposed throughout. ### Conclusion and open questions - Main empirical conclusions: - Current investment responds more slowly to real sales growth for firms facing higher measured uncertainty. - There is evidence of a nonlinear (convex) response of investment to real sales growth. - No clear long-run effect of uncertainty on capital accumulation is identified in this empirical work. - Outstanding question raised: - Whether the lack of identified long-run effect reflects limited time-series variation in the uncertainty measure or a deeper feature of investment behavior remains for future research. ### Appendix: data construction and tables overview - SIM survey details: - Annual survey of realized and planned investments among about 1,000 randomly selected manufacturing firms (population: businesses with more than 50 employees, energy sector excluded until 1999). - Unit: firm; stratified by firm size, geographical location, and sector (ATECO-91). - Managers interviewed; questionnaires sent end-December and collected by April of following year. - CADS details: - Centrale dei Bilanci dataset with roughly 800 items for about 40,000 firms; data available annually since 1981. - Capital stock estimation: - Perpetual inventory method with assumed depreciation rate  = 8 percent and benchmark capital stock on average three years old. - Additional tabulated items included in Appendix: - Table A1: Comparison of SIM and SIM-CADS samples. Total observations: 4,192 (SIM-CADS) vs. 14,873 (SIM). - Table A2: Balance of panel (final Firms count 564). - Table A3: Further descriptive statistics (selected): - Real sales growth (y_t): Median 0:0253, Mean 0:0204, Std. Dev. 0:1377 - Employment: Median 285, Mean 877, Std. Dev. 4091 - Observations per firm: Median 98:5:2:7 (as reported verbatim) ### References (selected headings from source) - Investment under uncertainty and irreversibility (authors and works listed). - Firm-level, plant-level, and microeconomic studies of investment (authors and works listed). - Econometric methods, panel data, and finite-sample issues (authors and works listed). - Financial frictions and capital markets (authors and works listed). - Data sources, surveys, and country-specific methodologies (authors and works listed). *Source: _wp04104 - Appendix Tables (excerpted material provided).* ### References.............................................................................................................. ### _wp04104 - References.............................................................................................................. ### References - References.................................................................................................................................26 ### Tables - 1. Frequency of Zero Investment and Disinvestment Episodes........................................11 - 2. Frequency of Zero Investment and Disinvestment Episodes........................................12 - 3. Investment and Expected Investment Rates .................................................................14 - 4. Firms with High and Low Uncertainty .........................................................................15 - 5. Investment Equations: One-Step Results......................................................................18 - 6. Investment Equations: Two-Step Results .....................................................................19 *References.................................................................................................................................26* ### Appendix Tables ### _wp04104 - Appendix Tables ### Introduction and main empirical findings - Aggregate investment exhibits large cyclical movements that can arise from firms' behavior under uncertainty rather than only from credit-market propagation mechanisms. - Using Italian manufacturing firms, evidence is found of: - Heterogeneity across firms in investment dynamics. - Slower adjustment of investment in response to demand shocks at higher levels of uncertainty. - An additional source of nonlinearity: a convex response of investment to demand shocks. - Policy implication highlighted: - A given demand stimulus will tend to have weaker short-run effects on investment when firms operate in a more uncertain environment. ### Theoretical background and predictions - Models of irreversible investment under uncertainty (real options) predict: - A wider inaction region under higher uncertainty, implying reduced short-run responsiveness of investment to demand shocks. - Threshold (trigger) policies: investment only when the gap between current and optimal capital is sufficiently large. - In multi-plant or multi-line settings with supermodularity, a convex response of firm-level investment to positive demand shocks is predicted. - Long-run effect of uncertainty on average capital stock is theoretically ambiguous (user-cost vs. hangover effects). ### Econometric specification and hypotheses tested - Baseline: error-correction model (ECM) derived from an ADL representation; dynamic panel framework. - Augmented ECM estimated (equation (5) in the source) includes: - Linear and quadratic real sales growth terms (y_it and (y_it)^2) to test for convexity/nonlinearity. - Interaction term (_it y_it) to test whether short-run response to demand shocks depends on measured uncertainty. - Long-run uncertainty term (^2_{i;t-1}) to test for persistent effects on capital stock level. - Control for cash flow (C_it / K_{i;t-1}) to avoid confounding liquidity effects. - Error structure: firm fixed effects (_i), time dummies (_t), and GMM-style instrumentation to address endogeneity. ### Data, sample, and stylized facts - Data sources: - Survey of Investment in Manufacturing (SIM) from Bank of Italy (annual, representative sample of manufacturing firms with more than 50 employees), electronic format since 1984. - Company Accounts Data Service (CADS) from Centrale dei Bilanci. - Sample construction and coverage: - Merged SIM and CADS; loss of approximately one-third of initial 14,873 observations. - Final sample: Observations 4,192; Firms 564. - Median firm size in SIM-CADS sample: 266 employees (Firm size (employees median)285266 in Table A1). - Period coverage: SIM data available 1984–1998; tables and analyses often use 1991–1998 or 1992–1995 for specific breakdowns. - Stylized facts on zero-investment episodes: - Frequency tables (Table 1 and Table 2) show higher incidence of zero investment for smaller firms and for disaggregated asset categories (buildings, plant and machinery, transportation). - Example entries from Table 1 (period 1991–1998, 2,682 observations): for Employees 50ñ99, Buildings Investment 53.30, Machinery Investment 4.95, Transport Investment 41.75, Total Investment 54.17, Total Disinvestment 45.52, Investment and Disinvestment 3.30. - Example entries from Table 2 (period 1992–1995, 1,689 observations): for Plants =1, Building Investment 38.96, Machinery Investment 2.19, Transport Investment 23.75, Total Investment 51.46, Total Disinvestment 35.57, Investment and Disinvestment 1.20. ### Measuring uncertainty - Firm-level uncertainty measure constructed from managers' one-year-ahead forecasts of their own investment: - Forecast error (f e)_it = I_it / K_{i;t-1} − E_{i;t-1}[I_it / K_{i;t-1}] (equation (6)). - Squared forecast error (f e)^2_it (equation (7)). - Rolling variance (time-varying) used as main measure: ^2_it = (1/t) Σ_{s=1}^t (f e)^2_is (equation (9)), estimated using at least 4 observations (t ≥ 4). - Descriptive relationships: - Table 3 (Investment and Expected Investment Rates): Sample mean I_it / K_{i;t-1} = 0.0962; Low Uncertainty mean = 0.1174; High Uncertainty mean = 0.0749. Sample mean E_{i;t-1}[I_it / K_{i;t-1}] = 0.0937. - Firms with higher measured uncertainty tend to invest less (and expect to invest less) in proportion to capital stock. - Table 4 shows sectoral distribution: e.g., Electrical goods Low Unc. 42:86, High Unc. 57:14; Machinery Low Unc. 46:32, High Unc. 53:68. Size: 50ñ99 employees Low Unc. 46:21, High Unc. 53:79; 100 or more employees Low Unc. 50:00, High Unc. 50:00. ### Estimation method and diagnostics - Dynamic panel estimation using GMM (system GMM: Arellano-Bover/Blundell-Bond), with both one-step and two-step estimators; Windmeijer (2000) finite sample correction applied to two-step standard errors. - Instruments: lagged levels and lagged differences of endogenous variables; current sales treated as predetermined in preferred specification. - Diagnostic tests reported: - Sargan (p) values reported in Tables 5 and 6 (columns): 0:34, 0:37, 0:46, 0:44, 0:34 (exactly as tabulated). - LM2 (p) (test for absence of 2nd-order serial correlation): 0:71, 0:83, 0:77, 0:76, 0:86 in one-step; 0:74, 0:78, 0:64, 0:64, 0:71 in two-step (as reported). ### Key empirical results (one-step and two-step GMM) - Sample size and firms: Observations 4;192; Firms 564 (reported in Tables 5 and 6). - One-step results (selected coefficients from Table 5): - Column (1) basic specification: - y_it: 0:1163 (std. err. 0:0588) - y_{i;t-1}: 0:1456 (0:0239) - (k−y)_{i;t−2}: −0:1354 (0:0273) - Column (2) includes quadratic term (y^2_it): - y^2_it: −0:3376 (0:1281) and later signs reported as positive in text; note table formatting shows coefficient labeled with leading minus sign for display. - Column (3) includes interaction ^2_it y_it: - ^2_it y_it: −0:9209 (0:3862) in one-step estimates (indicating weaker response at higher uncertainty). - Column (4) adds long-run uncertainty term: - ^2_{i;t-1}: −0:0199 (0:0545) (statistically small/insignificant). - Two-step results (selected coefficients from Table 6): - Column (1) basic specification: - y_it: 0:1030 (0:0255) - y_{i;t-1}: 0:1228 (0:0219) - (k−y)_{i;t−2}: −0:1314 (0:0228) - Column (2) quadratic term: - y^2_it: −0:2258 (0:1021) - Column (3) interaction: - ^2_it y_it: −0:8247 (0:3439) - Column (4) long-run uncertainty term: - ^2_{i;t−1}: −0:0115 (0:0477) - Interpretation emphasized in the source: - The interaction coefficient (one-step point estimate 0:92; two-step 0:8247) is large in absolute value and statistically different from zero, consistent with the theoretical prediction of weaker short-run responses under higher uncertainty. - Evidence of convexity (nonlinear response) is reported (authors note one-step point estimate 0:34 on quadratic term in text), supporting a convex response of investment to demand shocks. - Cash flow was not statistically significant in any specification. - Constant returns to scale hypothesis was not rejected and imposed throughout. ### Conclusion and open questions - Main empirical conclusions: - Current investment responds more slowly to real sales growth for firms facing higher measured uncertainty. - There is evidence of a nonlinear (convex) response of investment to real sales growth. - No clear long-run effect of uncertainty on capital accumulation is identified in this empirical work. - Outstanding question raised: - Whether the lack of identified long-run effect reflects limited time-series variation in the uncertainty measure or a deeper feature of investment behavior remains for future research. ### Appendix: data construction and tables overview - SIM survey details: - Annual survey of realized and planned investments among about 1,000 randomly selected manufacturing firms (population: businesses with more than 50 employees, energy sector excluded until 1999). - Unit: firm; stratified by firm size, geographical location, and sector (ATECO-91). - Managers interviewed; questionnaires sent end-December and collected by April of following year. - CADS details: - Centrale dei Bilanci dataset with roughly 800 items for about 40,000 firms; data available annually since 1981. - Capital stock estimation: - Perpetual inventory method with assumed depreciation rate  = 8 percent and benchmark capital stock on average three years old. - Additional tabulated items included in Appendix: - Table A1: Comparison of SIM and SIM-CADS samples (Abs. Freq., Rel. Freq. by sector, location, industrial sector, size, listed firms, firms in holding). Total observations: 4,192 (SIM-CADS) vs. 14,873 (SIM). - Table A2: Balance of panel (firm counts by start year and end year; final Firms count 564). - Table A3: Further descriptive statistics: - Real sales growth (y_t): Median 0:0253, Mean 0:0204, Std. Dev. 0:1377 - Employment: Median 285, Mean 877, Std. Dev. 4091 - Observations per firm: Median 98:5:2:7 (as reported verbatim) *Source: _wp04104 - Appendix Tables (excerpted material provided).* ### REFERENCES ### _wp04104 - REFERENCES ### Investment under uncertainty and irreversibility - Abel, Andrew B., 1983, ‘‘Optimal Investment under Uncertainty,’’ American Economic Review, Vol. 73, 228–233. - Abel, Andrew B., and Janine C. Eberly, 1996, ‘‘Optimal Investment with Costly Reversibility,’’ Review of Economic Studies, Vol. 63, 581–593. - Abel, Andrew B., and Janine C. Eberly, 1999, ‘‘The Effects of Irreversibility and Uncertainty on Capital Accumulation,’’ Journal of Monetary Economics, Vol. 44, 339–377. - Arrow, Kenneth J., 1968, ‘‘Optimal Capital Policy with Irreversible Investment,’’ in J. N. Wolfe (ed.), Value, Capital, and Growth, Papers in Honour of Sir John Hicks, Aldine Publishing Company, Chicago. - Bertola, Giuseppe, 1988, ‘‘Irreversible Investment,’’ Ph.D. Thesis, Massachusetts Institute of Technology, re-printed in Research in Economics 1998, Vol. 52, 3–37. - Caballero, Ricardo, 1991, ‘‘On the Sign of the Investment Uncertainty Relationship,’’ American Economic Review, Vol. 81, 279–288. - Caballero, Ricardo, 1999, ‘‘Aggregate Investment,’’ in John B. Taylor and Michael Woodford (eds.), Handbook of Macroeconomics, Vol.1, North-Holland, Amsterdam. - Dixit, Avinash (1997), ‘‘Investment and Employment Dynamics in the Short Run and the Long Run’’, Oxford Economic Papers, Vol. 49, 1–20. - Dixit, Avinash, and Robert Pindyck, 1994, Investment Under Uncertainty, Princeton University Press, Princeton, New Jersey. - Lucas, Robert E. Jr., and Edward C. Prescott, 1971, ‘‘Investment Under Uncertainty,’’ Econometrica, Vol. 39, 659–681. - Pindyck, Robert, 1988, ‘‘Irreversible Investment, Capacity Choice, and the Valuation of the Firm,’’ American Economic Review, Vol. 78, 969–985. - Hartman, Robert, 1972, ‘‘The Effects of Price and Cost Uncertainty on Investment,’’ Journal of Economic Theory, Vol. 5, 258–266. - Carruth, Alan, Andrew Dickerson, and Andrew Henley, 2000, ‘‘What Do We Know About Investment Under Uncertainty,’’ Journal of Economic Surveys, Vol. 2, 119–153. ### Firm-level, plant-level, and microeconomic studies of investment - Attanasio, Orazio, Isabel dos Reis, and Lia Pacelli, 2000, ‘‘Aggregate Implications of Plant Level Investment Behaviour: Evidence from the UK ARD,’’ mimeo, Institute for Fiscal Studies. - Caballero, Ricardo, Eduardo M. R. A. Engel, and John C. Haltiwanger, 1995, ‘‘Plant-Level Adjustment and Aggregate Investment Dynamics,’’ Brookings Papers on Economic Activity, Vol. 95, 1–54. - Doms, Mark, and Timothy Dunne, 1998, ‘‘Capital Adjustment Patterns in Manufacturing Plants,’’ Review of Economic Dynamics, Vol. 1, 409–429. - Bloom, Nicholas, 2000, ‘‘The Real Options Effect of Uncertainty on Investment and Labour Demand,’’ Institute for Fiscal Studies Working Paper no. 00/15. - Bloom, Nicholas, Stephen R. Bond, and John Van Reenen, 2003, ‘‘Uncertainty and Company Investment Dynamics: Empirical Evidence for UK Firms,’’ Centre for Economic Policy Research Discussion Paper no. 4025. - Mairesse, Jacques, Hall H. Bronwyn, and Benoit Mulkay, 1999, ‘‘Firm-Level Investment in France and the United States: an Exploration of What We Have Learned in Twenty Years,’’ NBER Working Paper no. 7437. - Nilsen, Anti O., and Fabio Schiantarelli, 2000, ‘‘Zeroes and Lumps in Investment: Empirical Evidence on Irreversibilities and Non-Convexities,’’ Economics Department Working Paper no. 337, Boston College. - Nucci, Francesco, and Alberto F. Pozzolo, 2001, ‘‘Investment and the Exchange Rate: an Analysis with Firm Level Data,’’ European Economic Review, Vol. 45, 259–283. - Bettoni, Andrea, 2000, ‘‘Capital Market Imperfections and Investment: Evidence from Firm Level Panel Data,’’ D.Phil. Thesis, University of Oxford. - Carpenter, Robert E., and Laura Rondi, 2000, ‘‘Italian Corporate Governance, Investment, and Finance,’’ CERIS-CNR Working Paper no. 14/2000. - Bond, Stephen R., Julie A. Elston, Jacques Mairesse, and Benoit Mulkay, 2003, ‘‘Financial Factors and Company Investment in Belgium, France, Germany and the UK,’’ Review of Economics and Statistics, Vol. 85, 153–165. - Bond, Stephen R., Dietmar Harhoff, and John Van Reenen, 1999, ‘‘Investment, R&D and Financing Constraints in Britain and Germany,’’ Institute for Fiscal Studies Working Paper no. 99/5. - Bond, Stephen R., and John Van Reenen, 2003, ‘‘Microeconometric Models of Investment and Employment,’’ mimeo, Institute for Fiscal Studies, London, available at http://www.ifs.org.uk/innovation/bondvanr.pdf. ### Econometric methods, panel data, and finite-sample issues - Arellano, Manuel, and Stephen Bond, 1991, ‘‘Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations,’’ Review of Economic Studies, Vol. 58, 277–297. - Blundell, Richard, and Stephen R. Bond, 1998, ‘‘Initial Conditions and Moment Restrictions in Dynamic Panel Data Models,’’ Journal of Econometrics, Vol. 87, 115–143. - Windmeijer, Frank, 2000, ‘‘A Finite Sample Correction for the Variance of Linear Two-Step GMM Estimators,’’ Institute for Fiscal Studies Working Paper no. 00/19. ### Financial frictions and capital markets - Bernanke, Ben S., Mark Gertler, and Simon Gilchrist, 1996, ‘‘The Financial Accelerator and the Flight to Quality,’’ The Review of Economics and Statistics, Vol. 78, 1–15. - Bernanke, Ben S., Mark Gertler, and Simon Gilchrist, 1998, ‘‘The Financial Accelerator in a Quantitative Business Cycle Framework,’’ NBER Working Paper no. 6455. - Hubbard, Glenn, 1998, ‘‘Capital-Market Imperfections and Investment,’’ Journal of Economic Literature, Vol. 36, 193–225. - Gelos, R. Gaston, and Alberto Isgut, 1999, ‘‘Fixed Capital Adjustment: Is Latin America Different? Evidence from the Colombian and Mexican Manufacturing Sectors,’’ International Monetary Fund Working Paper no. 99/59. - Pattillo, Catherine A., 1998, ‘‘Investment, Uncertainty, and Irreversibility in Ghana,’’ IMF Staff Papers, Vol. 45, 522–553. ### Data sources, surveys, and country-specific methodologies - Barca, Fabrizio, and others, 1996, ‘‘Metodi e Risultati dell’Indagine sugli Investimenti delle Imprese Industriali,’’ Supplementi al Bollettino Statistico, Banca d’Italia, Vol. 59. - Bigsten, Arne, and others, 1999, ‘‘Adjustment Costs, Irreversibility and Investment Patterns in African Manufacturing,’’ International Monetary Fund Working Paper no. 99/99. - Guiso, Luigi, and Giuseppe Parigi, 1999, ‘‘Investment and Demand Uncertainty,’’ Quarterly Journal of Economics, Vol. 114, 185–227. - Nickell, Stephen J., 1977a, ‘‘The Influence of Uncertainty on Investment,’’ Economic Journal, Vol. 86, 47–70. - Nickell, Stephen J., 1977b, ‘‘Uncertainty and Lags in the Investment Decision of Firms,’’ Review of Economic Studies, Vol. 48, 249–63. - Caballero, Ricardo, Eduardo M. R. A. Engel, and John C. Haltiwanger, 1995, ‘‘Plant-Level Adjustment and Aggregate Investment Dynamics,’’ Brookings Papers on Economic Activity, Vol. 95, 1–54. *Reference list from _wp04104 - REFERENCES* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp04104.pdf_