## _wp1049

## Source details

**Canonical URL:** [_wp1049](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp1049.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp1049.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp1049.pdf.json)

---

### I. Introduction — key themes and findings
- Persistent cross-country differences in output per worker are not primarily explained by variations in investment rates; evidence suggests total factor productivity (TFP) rather than capital accumulation accounts for observed per capita income differences (Hall and Jones, 1999).
- The substantial unexplained cross-country differences in TFP motivated a call for a “theory of TFP” (Prescott, 1998).
- Barriers to innovation, imitation, and adoption (for example, Parente and Prescott, 2000) and institutional factors (for example, Acemoglu and Johnson, 2005) have been advanced as explanations for low TFP in poorer countries.
- The recent global crisis and resulting uncertainty have increased concerns about growth prospects in many low-income countries.
- After a decade of almost-universally solid growth performance, the medium-term outlook for many low-income countries appears increasingly uncertain.
- If TFP drives growth, its determinants are the key to development prospects in low-income countries.
- Innovation operates through two channels:
  - Purposeful research and development that moves the global technological frontier (mainly in developed countries).
  - Adoption and adaptation of existing technology that helps countries converge toward the frontier.
- Innovation and financial services are strongly linked because invention and adoption are costly and risky and require financing.

### II. Empirical strategy and data
- Two-step agenda:
  - How do firm-specific actions (particularly innovation) affect firm productivity?
  - What is the role of the financial sector in facilitating higher productivity from innovation?
- Data and sample:
  - Firm-level data from the World Bank Enterprise Survey covering over 14,000 firms in 63 countries.
  - Countries include low, middle and high income countries; firms span all sizes with a focus on SME.
- Key sample statistics:
  - Richest country in sample: Ireland ($48,705).
  - Poorest country in sample: Burundi ($130).
  - Average GDP per capita in sample: $5263.
  - Regional firm shares: Latin America 37 percent; Eastern Europe and Central Asia 21 percent.
  - Industry shares: food 18 percent; metals and machinery 16 percent; garments 15 percent.
  - Self-reported financing obstacle distribution: 4 percent very severe obstacle; 7 percent major obstacle; 9 percent moderate obstacle; 15 percent minor; 65 percent no obstacle.
  - Ownership and trade: 2 percent government-owned; 11 percent foreign-owned; over 24 percent exporters.
  - Average firm age: 19 years (with some firms near 200 years).
  - Private credit to GDP in sample ranges from 3 percent (Guinea-Bissau) to 143 percent (Ireland).

### III. Main empirical findings
- Innovation and firm productivity
  - Firms that have introduced a new process or product are more productive, holding other factors equal.
  - Productive firms tend to be large, exporters, and privately owned by foreign or private domestic interests.
  - Results are robust across measures of innovation and productivity (output per worker and production-function based TFP).
  - Instrumenting for firm-level innovation (and using grouped averages as an alternative) preserves the positive effect of innovation on productivity.
  - Estimated impacts:
    - The estimated coefficient on New Technology implies a firm that adopts a new technology or develops a new product line is 7 percent more productive relative to a firm that does not engage in these innovative activities.
    - Using a Solow-residual measure of productivity, the estimated coefficient on New Technology implies an 8 percent higher productivity for innovators relative to non-innovators.
- Financial development mediates the innovation–productivity link
  - Hypothesis: financial sector development increases the effectiveness of innovation by better allocating capital to high-quality innovation projects.
  - Regression with interaction between average innovation and country-level financial development (mainly private credit to GDP) finds a positive and significant interaction term: innovation has a higher effect on productivity in financially developed countries.
  - Robustness across alternative financial development measures: stock market capitalization, financial openness, and a composite measure of access to financial services — interaction terms remain positive and significant.
  - Industry heterogeneity:
    - Interaction effects are larger and more significant for high-technology firms (metal and machinery, electronics, chemicals and pharmaceutics, auto components, other transport equipment) than for low-technology firms (food, beverages, garments, leather, textiles, wood and furniture, non-metallic and plastic materials, paper, other manufacture, agro-industry).
    - For high-tech firms, the positive mediation effect of financial development is particularly pronounced.

### IV. Key quantitative results (selected coefficients and sample statistics preserved exactly)
- Sample and summary statistics:
  - Total: 63 countries; 16,392 firms.
  - Firm-level: Age: 146401 / 9.00 / 18.00 / 1 / 190
  - Firm sales per worker (in '000 USD): 14640 / 57.43 / 552.77 / 0.00 / 325985.5
  - Size (number of employees): 14640 / 112.00 / 384.00 / 1 / 18000
  - GDP per capita (USD): 63 / 5,127 / 7,799 / 120 / 48705
  - Private Credit (percent of GDP): 63 / 0.35 / 0.31 / 0.02 / 1.43
- Baseline regression (Table 5) — dependent variable: log of output per worker:
  - Innovation coefficients:
    - Column (1): 0.156 (standard error 0.019) ***
    - Column (2): 0.112 (0.019) ***
    - Column (3): 0.093 (0.011) ***
    - Column (4): 0.066 (0.009) ***
    - Column (5): 0.162 (0.024) ***
    - Column (6): 0.110 (0.014) ***
  - Capacity utilization: 0.378 (0.049) ***
  - Age: 0.093 (0.012) ***
  - Small (firm size): −0.179 (0.021) ***
  - Exporter: 0.311 (0.025) ***
  - Foreign: 0.369 (0.034) ***
  - Observations examples: (1) 14640; R2 around 0.42.
- Aggregated innovation (Avg_Innovation) (Table 7):
  - Avg_Innovation coefficients:
    - Column (1): 0.257 (0.051) ***
    - Column (2): 0.171 (0.048) ***
    - Column (3): 0.138 (0.029) ***
    - Column (4): 0.067 (0.017) ***
    - Column (5): 0.289 (0.044) ***
    - Column (6): 0.518 (0.085) ***
  - Capacity utilization: 0.387 (0.049) ***
  - Observations examples: Column (1) 14640; R2 ≈ 0.41.
- Interaction: Avg_Innovation × Financial Development (Table 8) — dependent variable: log of output per worker:
  - Avg_Innovation X Fin. Dev:
    - Private credit: 2.706 (1.011) ***
    - Stock mkt capitalization: 7.361 (2.559) ***
    - Financial openness: 0.994 (0.348) ***
    - Access to financial services: 0.025 (0.009) ***
  - Capacity utilization: 0.412 (0.049) ***
  - Observations examples: Column (1) 14640; R2 ≈ 0.42.
- High-Tech vs Low-Tech industries (Table 9):
  - Innovation (New technology):
    - High-Tech industries: 0.147 (0.037) ***
    - Low-Tech industries: 0.154 (0.022) ***
  - Avg_Innovation X Private Credit:
    - High-Tech: 6.112 (2.356) ***
    - Low-Tech: 1.382 (1.1)
  - Observations: 43,441 / 10,296; R2 ≈ 0.41.
- TFP regressions (Tables 10–11):
  - Innovation on TFP:
    - Column (1) Innovation: 0.164 (0.023) ***
    - Column (2) New technology: 0.156 (0.025) ***
    - Column (3) New product: 0.109 (0.013) ***
    - Column (4) CoreIndex: 0.104 (0.013) ***
  - Interaction Avg_Innovation X Fin. Dev (TFP):
    - Private credit: 3.013 (0.984) ***
    - Stock mkt capitalization: 8.998 (2.991) ***
    - Financial openness: 1.208 (0.392) ***
    - Access to financial services: 0.028 (0.009) ***
  - Observations examples: 9,088; R2 ≈ 0.31–0.32.
- Robustness tests (Table 12):
  - Innovation (New technology) excluding three biggest countries:
    - Coefficient: 0.131 (0.020) ***
  - Avg_Innovation X Fin_Dev:
    - Coefficient: 3.327 (1.496) **
  - Observations: 108,293 / 151 (formatting preserved from source); R2 examples 0.46 and 0.71.

### V. Methodology and robustness
- Estimation framework:
  - Single-equation Cobb-Douglas-style log-linear specification estimated in levels, with industry and country dummies and firm controls (age, size, exporter status, ownership, share of skilled workers, capacity utilization or capital per worker).
  - Alternative two-step Solow-residual approach also used.
- Endogeneity and measurement strategies:
  - To address endogeneity and measurement error, innovation is sometimes measured as the average innovation by firms in a similar-sized location within a country (Avg_innovation).
  - Grouped averages (city-location averages) mitigate measurement error and two-way causality concerns; results remain significant and point estimates often increase.
- Robustness checks and findings:
  - Results robust to removing the three countries with largest samples (Bangladesh, India, Mexico): reduced sample of 60 countries and between 8000 and 11000 observations, interaction remains significant.
  - Results robust across country subsamples (low- and middle-income splits at GDP per capita $1000, exclusion of high-income countries) and across industry subsets.
  - Controlling for competition (as measured by firms’ pricing response to a 10 percent price increase) does not eliminate the significance of innovation; competition shows no direct effect on productivity in these specifications.
  - The Solow-residual based productivity measure (using a cost-of-capital assumption of 10 percent on reported net book value of assets) has a correlation of 0.96 with the paper’s innovation measure and yields similar conclusions.

### VI. Policy implications and conclusions
- Innovation broadly defined (new products, improved processes, adoption of existing technologies) has an important and measurable effect on firm performance.
- Financial market development is a key channel by which innovation raises productivity: in countries with well-developed financial markets, the returns to innovation are significantly larger.
- Policy priorities implied by the evidence:
  - Encourage adoption or imitation of existing technologies to increase TFP.
  - Continue the agenda of financial sector policy reform and strengthening, given the evidence that financial development enhances the productivity payoff to innovation.

### VII. Appendix highlights — Data variables and sources (selected definitions preserved exactly)
- GDP per capita: Logarithm of real GDP in USD, in current prices (as of the year of the survey). WEO
- Private credit to GDP: Private Credit by Deposit Money Banks divided by GDP. FSD
- Stock market capitalization: Stock Market Capitalization / GDP. FSD
- Financial Openness: Cross-border assets plus liabilities over GDP. MF
- Access to Financial Services: Composite measure of access to financial services. DB
- Average city innovation: Average value of the innovation measure in the country and city where the firm is located. Authors' computation
- TFPSolow: Solow residual from production function/cost share approach. Authors' computation
- Output per worker: Logarithm of sales divided by the average number of permanent workers over the last fiscal year. WBES
- Capacity Utilization: Average capacity utilization over last year (in %) (Capacity utilization is the amount of output actually produced relative to the maximum amount that could be produced with existing machinery and equipment and regular shifts). WBES
- New technology: Dummy that takes on the value of one if the firm has answered ‘yes’ to ‘Introduced new technology that has substantially changed the way that the main product is produced’. WBES
- New product: Dummy that takes on the value of one if the firm has answered ‘yes’ to ‘Developed a major new product line’. WBES
- Index7: A measure that adds one for every initiative the firm has undertaken in the three years before the survey from the menu of ‘Developed a major new product line’, ‘Upgraded an existing product line’, ‘Introduced new technology...’, ‘Agreed a new joint venture with foreign partner’, ‘Obtained a new licensing agreement’, ‘Outsourced a major production activity’ and ‘Brought in-house a major production activity’.
- Capital costs: Assumed to equal 10% of total assets. Authors' computation
- Abbreviations and data sources: DB, IFS, MF, FSD, WEO, WBES (as listed in source).

*Source: _wp1049 - REFERENCES*

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

### _wp1049 - References

### Tables and Appendix (listed items)
- Tables
  - 1. Sample of Countries
  - 2. Summary Statistics
  - 3. Distribution of Firms
  - 4. Correlation Matrix
  - 5. Baseline Regression
  - 6. Business Climate
  - 7. Baseline Regression with Aggregated Innovation
  - 8. Financial Development and Innovation: Interaction Effects
  - 9. Productivity, Innovation, and Financial Development: High Tech and Low Tech Industries
  - 10. Baseline Regression-TFP
  - 11. Financial Development and Innovation: Interaction Effects – TFP Measure
  - 12. Robustness Tests
- Appendix
  - 1. Data Variables and Sources

### I. INTRODUCTION — key themes and findings from the excerpt
- Persistent cross-country differences in output per worker are not primarily explained by variations in investment rates; evidence suggests total factor productivity (TFP) rather than capital accumulation accounts for observed per capita income differences (Hall and Jones, 1999).
- The substantial unexplained cross-country differences in TFP motivated a call for a “theory of TFP” (Prescott, 1998).
- Barriers to innovation, imitation, and adoption (for example, Parente and Prescott, 2000) and institutional factors (for example, Acemoglu and Johnson, 2005) have been advanced as explanations for low TFP in poorer countries.
- The recent global crisis and resulting uncertainty have increased concerns about growth prospects in many low-income countries.
- After a decade of almost-universally solid growth performance, the medium-term outlook for many low-income countries appears increasingly uncertain.
- The growth potential in many countries with low TFP remains dim, even beyond the eventual resolution of the crisis.

*Source: _wp1049 - References*

### conclusion of this crisis.

### _wp1049 - conclusion of this crisis.

### Key concepts: TFP, innovation, and financial development
- If TFP drives growth, its determinants are the key to development prospects in low-income countries.
- Some direct determinants of TFP—geography, history or institutions such as the prevalent legal system—are fixed or slow-moving; others, such as innovation or the level of financial development, can respond to external shocks, globalization or government policy.
- TFP is presented as a direct result of purposeful actions by economic agents and thus can be influenced by policy.
- Innovation operates through two channels:
  - Purposeful research and development that moves the global technological frontier (mainly in developed countries).
  - Adoption and adaptation of existing technology that helps countries converge toward the frontier.
- Innovation and financial services are strongly linked because invention and adoption are costly and risky and require financing.

### Empirical strategy and data
- Two-step agenda:
  - How do firm-specific actions (particularly innovation) affect firm productivity?
  - What is the role of the financial sector in facilitating higher productivity from innovation?
- Data and sample:
  - Firm-level data from the World Bank Enterprise Survey covering over 14,000 firms in 63 countries.
  - Countries include low, middle and high income countries; firms span all sizes with a focus on SME.
- Key sample statistics:
  - Richest country in sample: Ireland ($48,705).
  - Poorest country in sample: Burundi ($130).
  - Average GDP per capita in sample: $5263.
  - Regional firm shares: Latin America 37 percent; Eastern Europe and Central Asia 21 percent.
  - Industry shares: food 18 percent; metals and machinery 16 percent; garments 15 percent.
  - Self-reported financing obstacle distribution: 4 percent very severe obstacle; 7 percent major obstacle; 9 percent moderate obstacle; 15 percent minor; 65 percent no obstacle.
  - Ownership and trade: 2 percent government-owned; 11 percent foreign-owned; over 24 percent exporters.
  - Average firm age: 19 years (with some firms near 200 years).
  - Private credit to GDP in sample ranges from 3 percent (Guinea-Bissau) to 143 percent (Ireland).

### Main empirical findings
- Innovation and firm productivity
  - Firms that have introduced a new process or product are more productive, holding other factors equal.
  - Productive firms tend to be large, exporters, and privately owned by foreign or private domestic interests.
  - Results are robust across measures of innovation and productivity (output per worker and production-function based TFP).
  - Instrumenting for firm-level innovation (and using grouped averages as an alternative) preserves the positive effect of innovation on productivity.
  - Estimated impacts:
    - The estimated coefficient on New Technology implies a firm that adopts a new technology or develops a new product line is 7 percent more productive relative to a firm that does not engage in these innovative activities.
    - Using a Solow-residual measure of productivity, the estimated coefficient on New Technology implies an 8 percent higher productivity for innovators relative to non-innovators.
- Financial development mediates the innovation–productivity link
  - Hypothesis: financial sector development increases the effectiveness of innovation by better allocating capital to high-quality innovation projects.
  - Regression with interaction between average innovation and country-level financial development (mainly private credit to GDP) finds a positive and significant interaction term: innovation has a higher effect on productivity in financially developed countries.
  - Results are robust to alternative financial development measures: stock market capitalization, financial openness, and a composite measure of access to financial services — interaction terms remain positive and significant.
  - Industry heterogeneity:
    - Interaction effects are larger and more significant for high-technology firms (metal and machinery, electronics, chemicals and pharmaceutics, auto components, other transport equipment) than for low-technology firms (food, beverages, garments, leather, textiles, wood and furniture, non-metallic and plastic materials, paper, other manufacture, agro-industry).
    - For high-tech firms, the positive mediation effect of financial development is particularly pronounced.

### Methodology and robustness
- Estimation framework:
  - Single-equation Cobb-Douglas-style log-linear specification estimated in levels, with industry and country dummies and firm controls (age, size, exporter status, ownership, share of skilled workers, capacity utilization or capital per worker).
  - Alternative two-step Solow-residual approach also used.
  - To address endogeneity and measurement error, innovation is sometimes measured as the average innovation by firms in a similar-sized location within a country (Avg_innovation).
- Instrumenting and grouping:
  - Grouped averages (city-location averages) mitigate measurement error and two-way causality concerns; results remain significant and point estimates often increase.
- Robustness checks:
  - Results robust to removing the three countries with largest samples (Bangladesh, India, Mexico): reduced sample of 60 countries and between 8000 and 11000 observations, interaction remains significant.
  - Results robust across country subsamples (low- and middle-income splits at GDP per capita $1000, exclusion of high-income countries) and across industry subsets.
  - Controlling for competition (as measured by firms’ pricing response to a 10 percent price increase) does not eliminate the significance of innovation; competition shows no direct effect on productivity in these specifications.
  - The Solow-residual based productivity measure (using a cost-of-capital assumption of 10 percent on reported net book value of assets) has a correlation of 0.96 with the paper’s innovation measure and yields similar conclusions.

### Policy implications and conclusions
- Innovation broadly defined (new products, improved processes, adoption of existing technologies) has an important and measurable effect on firm performance.
- Financial market development is a key channel by which innovation raises productivity: in countries with well-developed financial markets, the returns to innovation are significantly larger.
- For growth and medium-term prospects in low-income countries, policies that:
  - Encourage adoption or imitation of existing technologies to increase TFP, and
  - Continue the agenda of financial sector policy reform and strengthening,
  are important priorities given the evidence that financial development enhances the productivity payoff to innovation.

*Source: _wp1049 - conclusion of this crisis.*

### REFERENCES

### _wp1049 - REFERENCES

### Key bibliographic themes
- Extensive literature on finance, innovation, and growth including works by:
  - Acemoglu and Johnson (2005); Aghion et al. (2005a, 2005b, 2005); Aghion, Howitt, and Mayer-Foulkes (2005).
  - Alfaro and Kalemli-Ozcan (2004); Angrist and Krueger (2001); Ayyagari, Demirgüç-Kunt, and Maksimovic (2007).
  - Beck (2008); Beck, Levine, and Loayza (2000); Beck, Demirgüc-Kunt, and Martinez Peria (2006); Beck, Demirgüc-Kunt, and Maksimovic (2005); Beck et al. (2008).
  - Coe, Helpman, and Hoffmaister (1997); Dollar, Hallward-Driemeier, and Mengistae (2005, 2006); Escribano and Guasch (2005).
  - Fisman and Love (2004); Fisman and Svensson (2007); Gatti and Love (2008).
  - Griliches and Mairesse (1997); Gorodnichenko, Svejnar, and Terrell (2008); Guiso, Sapienza, and Zingales (2004).
  - Haber (1997, 2003); Hall and Jones (1999); Hartmann et al. (2007); Hsieh and Klenow (2009).
  - Levine (2005); Maurer and Haber (2007); Parente and Prescott (2000); Prescott (1998); Rajan and Zingales (1998); Raddatz (2006); Sharma (2007).

### Sample of countries (Table 1)
- Total: 63 countries; 16,392 firms.
- Selected country entries (Country — GDP per capita (in USD) — Private Credit/ GDP — Number of firms):
  - Angola 2,8480.06207 (format from source)
  - Argentina 54580.11608
  - Armenia 14770.07202
  - Burundi 1200.18102
  - Bangladesh 4150.331200
  - Bulgaria 41200.43511
  - Bosnia and Herzegovina 27510.3940
  - Belarus 30970.0937
  - Bolivia 11670.35304
  - Botswana 70210.2110
  - Chile 89030.62587
  - Cameroon 9790.0969
  - Colombia 29110.22602
  - Cape Verde 24250.4124
  - Costa Rica 46670.32295
  - Czech Republic 121910.3355
  - Ecuador 30580.22315
  - Spain 260771.3105
  - Egypt 11370.51944
  - Estonia 103430.4633
  - Georgia 14840.1127
  - Guinea-Bissau 1900.0350
  - Greece 222900.7876
  - Guatemala 23270.25295
  - Honduras 14620.42240
  - Hungary 109440.48273
  - India 7170.371814
  - Ireland 487051.43155
  - Jamaica 35320.2340
  - Jordan 25190.88333
  - Kazakhstan 37860.28245
  - Kyrgyzstan 4790.0746
  - Korea 164440.89182
  - Lebanon 61471.8887
  - Lithuania 75360.2939
  - Latvia 69550.4826
  - Moldova 8830.22118
  - Madagascar 3090.09216
  - Mexico 80600.17957
  - Macedonia 28600.2427
  - Mauritania 9380.2479
  - Mauritius 49720.73136
  - Malawi 2220.06145
  - Namibia 33890.5799
  - Niger 2650.067
  - Nicaragua 8960.24324
  - Panama 52170.74184
  - Peru 33660.17342
  - Poland 79650.26391
  - Portugal 175871.41113
  - Paraguay 16570.16296
  - Romania 44530.17334
  - Russia 53260.2387
  - Rwanda 3120.1159
  - Serbia and Montenegro 35260.2752
  - El Salvador 26610.42416
  - Slovakia 88540.3229
  - Slovenia 175590.5153
  - Swaziland 24310.1864
  - Tajikistan 3640.1655
  - Turkey 71100.2493
  - Tanzania 3720.11270
  - Uganda 3180.06304
  - Ukraine 18420.15135
  - Uruguay 60360.25285
  - Uzbekistan 547.58
  - Vietnam 6380.59237
  - Dem. Republic of Congo 1470.02149

(Note: "Number of firmsCountryGDP per capita (in USD)Private Credit/ GDP" formatting preserved from source tables.)

### Summary statistics (Table 2)
- Firm-level variables (Obs / Mean / Std. Dev. / Min / Max where available):
  - Age: 146401 / 9.00 / 18.00 / 1 / 190
  - Firm sales per worker (in '000 USD): 14640 / 57.43 / 552.77 / 0.00 / 325985.5
  - Size (number of employees): 14640 / 112.00 / 384.00 / 1 / 18000
  - Government ownership: 14640 / 0.02 / 0.14 / 0 / 1
  - Foreign ownership: 14640 / 0.11 / 0.31 / 0 / 1
  - Exporter: 14640 / 0.24 / 0.43 / 0 / 1
  - Access to finance: 14385 / 0.68 / 1.08 / 0 / 4
  - New technology: 14640 / 0.47 / 0.50 / 0 / 1
  - New product: 14625 / 0.51 / 0.50 / 0 / 1
  - Share of skilled production workers: 14640 / 63.98 / 34.42 / 0 / 100
- Macroeconomic variables:
  - GDP per capita (USD): 63 / 5,127 / 7,799 / 120 / 48705
  - Private Credit (percent of GDP): 63 / 0.35 / 0.31 / 0.02 / 1.43
  - Stock market capitalization (percent of GDP): 47 / 0.27 / 0.23 / 0.00 / 11.07
  - Financial openness: 59 / 1.84 / 2.37 / 0.39 / 18.8

### Distribution of firms (Table 3)
- Industry shares (Number of firms / Share of firms (%)):
  - Textiles 1180 / 8.1
  - Leather 351 / 2.4
  - Garments 2301 / 15.7
  - Food 2672 / 18.3
  - Beverages 588 / 4.0
  - Metals and machinery 2300 / 15.7
  - Electronics 357 / 2.4
  - Chemicals and pharmaceutics 1499 / 10.2
  - Wood and furniture 677 / 4.6
  - Non-metallic and plastic materials 733 / 5.0
  - Paper 251 / 1.7
  - Other manufacturing 1543 / 10.5
  - Auto and auto components 188 / 1.3
  - Total 14640 / 100.0
- Size distribution:
  - Small (below 20 employees) 6634 / 45.3
  - Medium (20-100 employees) 4945 / 33.8
  - Large (more than 100 employees) 3061 / 20.9
  - Total 14640 / 100.0
- Location distribution:
  - Capital City 6795 / 46.4
  - Other city with population >1 Million 2803 / 19.2
  - City with population 250,000 - 1,000,000 2026 / 13.8
  - City with population 50,000-250,000 1473 / 10.1
  - City with population of less than 50,000 1543 / 10.5
  - Total 14640 / 100

### Correlation matrix highlights (Table 4)
- Firm-level (selected):
  - Correlation New product — New technology: 0.34881
  - Correlation Output per worker — New product: 0.1544
  - Correlation Business constraint: Cost of finance — Business constraint: Access to finance: 0.75811
- Country-level:
  - GDP per capita (USD) — Private Credit (percent of GDP): 0.60371
  - GDP per capita (USD) — Financial openness: 0.8074
  - Private Credit (percent of GDP) — Stock market capitalization (percent of GDP): 0.53761

### Main regression findings (selected coefficients preserved exactly)

- Baseline regression (Table 5) — dependent variable: log of output per worker; industry, country, and city-size dummies included:
  - Innovation coefficients:
    - Column (1): 0.156 (standard error 0.019) ***
    - Column (2): 0.112 (0.019) ***
    - Column (3): 0.093 (0.011) ***
    - Column (4): 0.066 (0.009) ***
    - Column (5): 0.162 (0.024) ***
    - Column (6): 0.110 (0.014) ***
  - Capacity utilization: 0.378 (0.049) ***
  - Age: 0.093 (0.012) ***
  - Small (firm size): −0.179 (0.021) ***
  - Exporter: 0.311 (0.025) ***
  - Foreign: 0.369 (0.034) ***
  - Observations vary by column; e.g., (1) 14640; R2 around 0.42.

- Business climate (Table 6) — dependent variable: log of output per worker:
  - Innovation: coefficients range from 0.152 to 0.160 across columns; e.g., Column (1): 0.157 (0.019) ***
  - Business constraint: Access to finance: −0.035 (0.010) ***
  - Business constraint: cost of finance: −0.020 (0.012) *
  - Capacity utilization: 0.346 (0.049) ***
  - Observations examples: Column (1) 14385; R2 ranges 0.41–0.45.

- Aggregated innovation (Table 7) — Avg_Innovation:
  - Avg_Innovation coefficients:
    - Column (1): 0.257 (0.051) ***
    - Column (2): 0.171 (0.048) ***
    - Column (3): 0.138 (0.029) ***
    - Column (4): 0.067 (0.017) ***
    - Column (5): 0.289 (0.044) ***
    - Column (6): 0.518 (0.085) ***
  - Capacity utilization: 0.387 (0.049) ***
  - Observations examples: Column (1) 14640; R2 ≈ 0.41.

- Interaction: Financial development × Innovation (Table 8) — dependent variable: log of output per worker:
  - Avg_Innovation X Fin. Dev:
    - Private credit: 2.706 (1.011) ***
    - Stock mkt capitalization: 7.361 (2.559) ***
    - Financial openness: 0.994 (0.348) ***
    - Access to financial services: 0.025 (0.009) ***
  - Capacity utilization: 0.412 (0.049) ***
  - Observations examples: Column (1) 14640; R2 ≈ 0.42.

- High-Tech vs Low-Tech industries (Table 9):
  - Innovation (New technology):
    - High-Tech industries: 0.147 (0.037) ***
    - Low-Tech industries: 0.154 (0.022) ***
  - Avg_Innovation X Private Credit:
    - High-Tech: 6.112 (2.356) ***
    - Low-Tech: 1.382 (1.1) (standard error formatting from source)
  - Observations: 43,441 / 10,296; R2 ≈ 0.41.

- TFP dependent regressions (Table 10):
  - Innovation on TFP:
    - Column (1) Innovation: 0.164 (0.023) ***
    - Column (2) New technology: 0.156 (0.025) ***
    - Column (3) New product: 0.109 (0.013) ***
    - Column (4) CoreIndex: 0.104 (0.013) ***
  - Observations: examples 9,088; R2 ≈ 0.31.

- Interaction effects with TFP (Table 11):
  - Avg_Innovation X Fin. Dev:
    - Private credit: 3.013 (0.984) ***
    - Stock mkt capitalization: 8.998 (2.991) ***
    - Financial openness: 1.208 (0.392) ***
    - Access to financial services: 0.028 (0.009) ***
  - Capacity utilization: 0.502 (0.059) ***
  - Observations examples: 9,088; R2 ≈ 0.32.

- Robustness tests (Table 12):
  - Innovation (New technology) excluding three biggest countries:
    - Coefficient: 0.131 (0.020) ***
  - Avg_Innovation X Fin_Dev:
    - Coefficient: 3.327 (1.496) **
  - Observations: 108,293 / 151, (source formatting preserved); R2 examples 0.46 and 0.71.

### Empirical interpretation (from tables)
- Firm-level innovation measures (New product, New technology, composite indices, and local averages) are positively and significantly associated with higher log output per worker and higher TFP across multiple specifications.
- Capacity utilization and exporter status consistently show large positive associations with output per worker and TFP.
- Small firm status is negatively associated with output per worker and TFP.
- Interaction terms indicate that higher financial development (private credit, stock market capitalization, financial openness, access to financial services) amplifies the positive relationship between local/aggregate innovation and firm-level productivity.
- Business constraints related to access to finance and cost of finance have negative associations with output per worker.

*Source: _wp1049 - REFERENCES (tables and reference list as provided in the source PDF).*

### Appendix 1. Data Variables and Sources

### Appendix 1. Data Variables and Sources

### Variable definitions (macroeconomic and country-level)
- GDP per capita  
  - Logarithm of  real GDP in USD, in current prices (as of  the year of  the survey). WEO
- Private credit to GDP  
  - Private Credit by Deposit Money Banks divided by GDP. FSD
- Stock market capitalization  
  - Stock Market Capitalization / GDP. FSD
- Financial Openness  
  - Cross-border assets plus liabilities over GDP. MF
- Access to Financial Services  
  - Composite measure of  access to f inancial services. DB
- Average city innovation  
  - Average value of  the innovation measure in the country and city w here the f irm is located. Authors' computation
- TFPSolow  
  - Solow  residual f rom production f unction/cost share approach. Authors' computation
- Exchange rate  
  - Exchange rate: LCU per USD, end of  period, period average (period ref ers to the survey year). IFS

### Variable definitions (firm-level: performance, characteristics, and controls)
- Sales  
  - Total sales, converted f rom LCU into USD using the average exchange rate. WBES
- Output per w orker  
  - Logarithm of  sales divided by the average number of  permanent w orkers over the last f iscal year. WBES
- Age  
  - Logarithm of  year of  the survey minus year the f irm started operations. WBES
- Capac ity  Utiliz ation  
  - Average capacity utilization over last year (in %) (Capacity utilization is the amount of  output actually produced relative to the maximum amount that could be produced w ith existing machinery and equipment and regular shif ts). WBES
- Small  
  - Dummy that takes on the value of  one if  the f irm has less than 20 permanent employees. WBES
- Medium  
  - Dummy that takes on the value of  one if  the f irm has betw een 20 and 100 permanent employees. WBES
- Large  
  - Dummy that takes on the value of  one if  the f irm has more than 100 employees. WBES
- Exporter  
  - Dummy that takes on the value of  one if  the f irm generates any sales f rom exports. WBES
- Government  
  - Dummy that takes on the value of  one if  the government ow ns a share of  the f irm. WBES
- Foreign ow ned  
  - Dummy that takes on the value of  one if  a f oreign entity ow ns a share of  the f irm. WBES
- New  technology  
  - Dummy that takes on the value of  one if  the f irm has answ ered ‘yes’ to ‘Introduced new  technology that has substantially changed the w ay that the main product is produced’. WBES
- New  product  
  - Dummy that takes on the value of  one if  the f irm has answ ered ‘yes’ to ‘Developed a major new  product line’. WBES
- Cor e  
  - Sum of  ‘New  technology’ and ‘New  product’.
- Index7  
  - A measure that adds one f or every initiative the f irm has undertaken in the three years bef ore the survey f rom the menu of  ‘Developed a major new  product line’, ‘Upgraded an existing product line’, ‘Introduced new  technology...’, ‘Agreed a new  joint venture w ith f oreign partner’, ‘Obtained a new  licensing agreement’, ‘Outsourced a major production activity’ and ‘Brought in-house a major production activity’.
- Obstacle: Access to Finance  
  - Obstacle variable: a value of  0 indicates no problem, values of  1-4 indicate a slight to major problem, respectively. WBES
- Obs tac le: Elec tric ity  
  - Obstacle variable: a value of  0 indicates no problem, values of  1-4 indicate a slight to major problem, respectively. WBES
- Obstacle: Transportation  
  - Obstacle variable: a value of  0 indicates no problem, values of  1-4 indicate a slight to major problem, respectively. WBES
- Obstacle: Business Licensing and Operating Permits  
  - Obstacle variable: a value of  0 indicates no problem, values of  1-4 indicate a slight to major problem, respectively. WBES
- Obstacle: Crime, Thef t and Disorder  
  - Obstacle variable: a value of  0 indicates no problem, values of  1-4 indicate a slight to major problem, respectively. WBES
- Obstacle: Anti-Competitive and Inf ormal Practices  
  - Obstacle variable: a value of  0 indicates no problem, values of  1-4 indicate a slight to major problem, respectively. WBES

### Variable definitions (firm-level: balance sheet and cost measures)
- Capital  
  - Net book value of  total assets of  the f irm, w here necessary converted f rom LCU into USD. WBES
- Labor  
  - Average number of  w orkers over last f iscal year. WBES
- Compensation  
  - Total compensation OR Manpow er costs, w here necessary converted f rom LCU into USD. WBES
- Capital costs  
  - Assumed to equal 10% of  total assets. Authors' computation

### Other firm-level controls and coding
- City size  
  - A coded variable taking on the values of  1 (Capital City), 2 (other city > 1 Mill.), 3 (250K – 1 Mill.) 4 (50K-250K) and 5 (<50K) C38. WBES
- Country s pe cific variable s  
  - (Listed as a category header in source)
- Firm  s pe cific variable s  
  - (Listed as a category header in source)

### Abbreviations and data sources
- DB: Finance for All? Policies and Pitfalls in Expanding Access, World Bank, November 13, 2007, Asli Demirgüç-Kunt and Thorsten Beck http://go.worldbank.org/S3EWEOI440
- IFS: International Financial Statistics
- MF: Philip R. Lane and Gian Maria Milesi-Ferretti (2006) "The External Wealth of Nations Mark II: Revised and Extended Estimates of Foreign Assets and Liabilities, 1970-2004", IMF Working Paper 06/69
- FSD: Financial Structure Dataset  
  - Thorsten Beck, Asli Demirgüç-Kunt and Ross Levine, (2000), "A New Database on Financial Development and Structure," World Bank Economic Review 14, 597-605
- WEO: World Economic Outlook
- WBES: World Bank Enterprise Survey

*Source: Appendix 1. Data Variables and Sources*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2010/_wp1049.pdf_