## _wp0894 - References

## Source details

**Canonical URL:** [_wp0894 - References](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp0894.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp0894.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp0894.pdf.json)

---

### Introduction: motivation, goal, and headline findings
- Motivation:
  - Firms’ incentives to operate informally are shaped by the possibility of reducing or eliminating tax payments, and avoiding burdensome and often complex administrative procedures associated with tax and regulatory compliance and bureaucratic corruption.
  - Operating informally may limit firms’ access to bank financing because banks tend to be unwilling to grant credit to enterprises that lack proper documentation; hidden assets generally cannot be used as collateral.
  - Deficiencies in the legal and institutional environment can discourage firms from seeking bank financing and make it difficult for banks to enforce contracts, sort and monitor borrowers, resulting in credit rationing.
  - Access to bank finance has a significant and positive effect on firm performance.

- Goal of the paper:
  - Examine the relationship between informality and bank credit and the role of tax and legal institutions in determining this.
  - Develop a simple model in which firms choose informality; both the extent of informality and a firm’s reliance on bank credit are related to policy and institutional constraints.
  - Test predictions using a firm-level data set for 26 countries in Eastern Europe and Central Asia.

- Key empirical findings (summary):
  - Strong negative association between informality and bank credit.
  - A firm’s reliance on bank credit is positively associated with the quality of the legal environment and negatively associated with weaknesses in tax administration (proxied by bribes to tax collectors and severity of tax administration constraints).
  - Weaknesses in tax administration are positively associated with use of informal credit and with the probability that firms self-report financing availability as a major obstacle.
  - Country-level: firms use less bank credit in countries with underdeveloped financial systems and where tax compliance costs are high.
  - Robustness: negative association robust to inclusion of firm transparency and performance controls and to instrumental variables addressing endogeneity.

### Analytical framework (model): setup and propositions
- Timing and production:
  - Representative firm operates for two periods i = 1,2.
  - Production yi = f(ki , ), with 0(.)'>f and 0(.)''<f.
  - Initial capital k1; in period 2 financing uses retained earnings π1 and external borrowing L; 2k = k1 + π1 + L (displayed as 21 kLπ=+ in source).
  - Depreciation assumed to be 1; firm dissolved in period 2.

- Informality choice and payoffs:
  - Firm can hide fraction αi of activity (α = 0 full formality, α = 1 complete evasion).
  - Formal operations reported revenue subject to burden rate T.
  - Expected penalty P = C(α)y with C(.) convex and increasing in α and y.
  - Profits:
    - Period 1: π1 = (1 − α1)(1 − T)y1 − C(α1)y1 (source notation: 11111 (1   (1)   )()Ty  Cyπαα=−−− ).
    - Period 2: π2 = (1 − α2)(1 − T)y2 − C(α2)y2 + (1 − α2)Lr (source notation: 22222 (1   (1)   )()(1)    Ty   CyrLπαα=−−−−+ ).

- Credit ceiling and comparative statics:
  - Credit ceiling L = L(α1; k1, q) with:
    - ∂L/∂α < 0 (credit ceiling decreasing in firm informality),
    - ∂L/∂k1 > 0,
    - ∂L/∂q > 0 (q = quality of legal and financial institutions).
  - Two cases:
    - Credit ceiling not binding: L(α1; k1, q) > L. Then firm’s share of informal operations is decreasing in tax/regulatory burden with sign dαi/dT = 1/C′′(αi) > 0 (i = 1,2) (as stated in source).
    - Credit ceiling binding: L(α1; k1, q) = L.
  - Proposition 1: A larger tax burden discourages bank financing when the disincentive for investment and the sensitivity of α with respect to the tax burden are sufficiently large.
  - Proposition 2: A better quality of legal and financial institutions increases bank financing.

### Empirical approach, data, and variable definitions
- Data:
  - 2005 Business Environment and Enterprise Performance Survey (BEEPS).
  - Sample: firms in 26 transition countries in Eastern Europe and Central Asia.
  - Analysis restricted to firms that have external financing sources.
  - Total observations: 4,226 firms.

- Dependent binary indicators (definitions from source):
  - Bank Credit: 1 if firm relies on external sources (local and foreign banks) to finance working capital or new investment, 0 otherwise.
  - Informal Credit: 1 if firm relies on money-lenders or family and friends for working capital or new investment, 0 otherwise.
  - Financing Obstacle: 1 if firm reports availability of financing as a “major constraint” (rating of 4 on 1–4 scale), 0 otherwise.
  - Informality: share of firm's operations conducted informally (percentage of sales not reported for tax purposes).

- Key descriptive statistics (exact values from source):
  - Over 56 percent of firms use bank credit.
  - 29 percent use informal credit sources.
  - 22 percent report availability of financing as a major obstacle.
  - Sales to tax authorities are not fully reported by 39 percent of firms; average level of informality is 11 percent.
  - Selected sample averages and ranges:
    - Bank Credit: Obs 4226, Mean 0.56, Std. Dev. 0.50, Min 0, Max 1
    - Informal Credit: Obs 4226, Mean 0.29, Std. Dev. 0.45, Min 0, Max 1
    - Financing Obstacle: Obs 4085, Mean 0.22, Std. Dev. 0.41, Min 0, Max 1
    - Informality: Obs 4226, Mean 11.20, Std. Dev. 19.24, Min 0, Max 98
    - Index of legal quality: Obs 3626, Mean 0.91, Std. Dev. 0.63, Min 0, Max 2
    - Private Credit (percent of GDP): Obs 4226, Mean 29.64, Std. Dev. 15.39, Min 6.19, Max 59
    - Time to pay taxes (hours per year): Obs 4226, Mean 486.82, Std. Dev. 547.82, Min 9, Max 2185
    - GDP per capita (log): Obs 4226, Mean 7.82, Std. Dev. 0.86, Min 5.47, Max 9.34
    - Inflation (annual percent): Obs 4226, Mean 5.84, Std. Dev. 4.31, Min 0.51, Max 7.30
    - Real GDP growth (percent): Obs 4226, Mean 4.99, Std. Dev. 2.81, Min 0.11, Max 4.55
    - Rule of law: Obs 4226, Mean -0.17, Std. Dev. 0.71, Min -1.32, Max 1

- Estimation:
  - Reduced-form Probit models for 0–1 dependent variables with heteroskedasticity-robust standard errors.
  - Controls: firm age, size, industry dummies, government ownership, exporter dummy, corporation dummy; country controls include log real GDP per capita, CPI inflation, real GDP growth.
  - Economic effects evaluated at the mean of controls.

### Main empirical findings (detailed results and magnitudes)
- Informality and bank finance (baseline):
  - Informality is statistically significantly and negatively associated with reliance on bank financing.
  - Marginal effect: probability of relying on bank credit increases by 2 percentage points for every 10 percent reduction in informality at the sample average of 11 percent.
  - Informality is positively and significantly associated with reliance on informal credit and with higher incidence of severe financing constraints.

- Firm size and financing:
  - Probability of relying on bank credit decreases by 27 percent for small firms compared to large firms (marginal effect reported).
  - Reliance on informal sources increases by 13 percent for small firms.

- Business environment effects:
  - Weaknesses in tax administration negatively and significantly affect reliance on bank credit.
    - Example: a firm reporting frequent bribes to deal with taxes has a 5 percent lower probability of using bank credit (column 4 marginal effect).
  - Legal quality positively associated with reliance on bank credit.
    - Improvement in legal quality index from sample average 0.91 associated with a 4-percent increase in probability of using bank credit (estimate implication).
  - Tax administration weaknesses positively associated with use of informal credit and higher probability of reporting financing as a major obstacle:
    - A firm that sees tax administration as a major obstacle has a 6 percent higher probability of using informal credit and a 12 percent higher probability of reporting being financially constrained (columns 8 and 12 estimates).

- Formal versus informal firms:
  - Informal firms benefit more from improvements in legal system quality: improvement in legal quality index results in a 7 percent increase in probability of informal firms using bank credit.
  - Frequency of bribes reduces bank credit use by formal firms; tax administration constraints reduce bank credit use by informal firms.
  - Weak tax administration results in an 8 percent increase in probability informal firms rely on informal financing and a 4 percent increase for formal firms.

- Country-level institutional effects:
  - More developed financial systems (private credit/GDP) and better rule of law associated with higher use of bank credit.
  - Higher tax compliance costs (time to pay taxes) negatively associated with reliance on bank credit.
    - Example: if Ukraine took the same number of hours per year to comply with taxes as Estonia (104) instead of Ukraine’s level (2185), there would be a 4-percentage point increase in the probability that Ukrainian firms rely on bank credit (using column 4 results).
  - Firms in richer countries with lower inflation rely more on bank credit; GDP growth is insignificant in bank credit regressions.
  - Poorer and faster growing countries tend to rely more on informal financing.

- Interaction effects:
  - Informal firms’ reliance on bank credit is most severely affected by weaknesses in the tax system.
  - Interaction coefficients of informality with financial sector development and rule of law are insignificant.
  - Supports model prediction: where effective tax burden is high, informal firms have greater incentive to forego bank credit.

- Firm transparency and performance controls:
  - External auditing is positively and significantly associated with reliance on bank credit.
  - Use of international accounting standards (IAS) not significantly associated with bank credit in baseline; however interactions show the negative relationship between informality and bank credit is less likely when firm uses international accounting standards.
  - Past sales growth predicts bank financing; larger fixed assets increase probability of bank finance and reduce reliance on informal credit.
  - Informality results robust to inclusion of auditing, IAS, sales growth, and assets.

- Instrumental variables (endogeneity) results:
  - Instrument: average informality in firms within the same location (correlation with firm-level informality = 0.42).
  - First-stage F-test indicates the instrument is strong.
  - IV estimates: instrumented informality remains negative and significant for use of bank credit and perceived financing obstacles; positive and significant for use of informal credit.
  - Reported IV coefficients (selected):
    - Informality coefficient in Bank Credit IV regression: -0.005 (standard error 0.003, significant at 10 percent in column shown).
    - Average Informality first-stage coefficient: 1.039 (standard error 0.038, significant at 1 percent).

### Policy implications and recommendations (from analysis and results)
- Institutional quality and legal environment:
  - Strengthen the legal environment to reduce informality and encourage bank borrowing even for financially opaque informal firms.
  - Improve rule of law and contract enforcement to increase bank financing availability.

- Tax policy and administration:
  - Reduce tax compliance burdens and streamline regulations to lower incentives for informality.
  - Strengthen tax administration and reduce corrupt practices (bribes) to increase formal-sector borrowing and reduce reliance on informal credit.
  - High tax compliance costs amplify the negative association between informality and bank credit; lowering compliance time (hours) can increase bank credit use.

- Financial sector development and firm transparency:
  - Deepen financial systems (increase private credit/GDP) to raise the use of bank credit.
  - Encourage external auditing and adoption of international accounting standards to mitigate negative effects of informality on access to bank finance.

- Broader conclusion:
  - Policies that reduce tax and regulatory constraints and improve legal and financial institutions can reduce the size of the informal economy and promote financial deepening by increasing benefits and lowering costs of participating in formal credit markets.

*Italic source attribution: Content derived from _wp0894 - References.*

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

### _wp0894 - References

### Introduction
- Motivation:
  - Firms’ incentives to operate informally are shaped by the possibility of reducing or eliminating tax payments, and avoiding burdensome and often complex administrative procedures associated with tax and regulatory compliance and bureaucratic corruption (see Dabla-Norris et. al, 2008 and references therein).
  - Operating informally may limit firms’ access to bank financing because banks tend to be unwilling to grant credit to enterprises that lack proper documentation (government registration and licensing, tax compliance certificates, audited financial statements); hidden assets generally cannot be used as collateral; inaccurate representation of firms’ financial soundness and economic prospects can deter lenders (World Bank, 2007a).
  - Deficiencies in the legal and institutional environment can discourage firms from seeking bank financing and make it difficult for banks to enforce contracts, sort and monitor borrowers, resulting in credit rationing (Demirgüç-Kunt and Maksimovich, 1998; Beck et. al, 2005; Savafian and Wimpey, 2007).
  - Access to bank finance has a significant and positive effect on firm performance (Klapper et. al, 2006; Claessans and Leavan, 2003; Aiyagari et. al, 2007).

- Goal of the paper:
  - Examine the relationship between informality and bank credit and the role of tax and legal institutions in determining this.
  - Develop a simple model in which firms choose informality; both the extent of informality and a firm’s reliance on bank credit are related to policy and institutional constraints.
  - Test predictions using a firm-level data set for 26 countries in Eastern Europe and Central Asia.

- Key empirical findings summarized:
  - Strong negative association between informality and bank credit.
  - A firm’s reliance on bank credit is positively associated with the quality of the legal environment and negatively associated with weaknesses in tax administration (proxied by bribes to tax collectors and severity of tax administration constraints).
  - Weaknesses in tax administration are positively associated with use of informal credit and with the probability that firms self-report financing availability as a major obstacle.
  - Differential impacts by firm formality status:
    - A well-functioning legal system increases access to bank credit even for financially opaque informal firms while lowering reliance on informal credit by formal firms.
    - Weak tax administration increases reliance on informal credit by informal firms more than for formal firms.
  - Country-level findings:
    - Firms use less bank credit in countries with underdeveloped financial systems and where tax compliance costs are high.
    - An interaction term between tax compliance costs and the level of informal activity is negative and significant, implying the negative association between informality and credit is stronger where tax administration is weak.
  - Robustness and endogeneity:
    - The negative association is robust to inclusion of variables capturing firm transparency and performance.
    - Instrumental variables techniques addressing potential endogeneity between informality and bank credit leave results unchanged.

### Analytical framework (model)
- Setting and timing:
  - Representative firm operates for two periods i = 1,2 in a competitive environment.
  - Production: yi = f(ki , ), with 0(.)'>f and 0(.)''<f.
  - Initial capital in period 1 is k1.
  - In period 2, firm finances capital with retained earnings π1 and external borrowing L; 2k = k1 + π1 + L (equation displayed as 21 kLπ=+ in source).
  - Rate of capital depreciation assumed to be 1.
  - Firm dissolved in period 2; dividend for period 1 assumed zero.
  - Financial intermediary: one-period risk-free security earns net return r > 0. (Note: model interpreted as small open economy.)

- Informality choice and costs:
  - In each period firm can operate formally or hide fraction αi of activity; α = 0 denotes full formality, α = 1 denotes complete evasion.
  - Formal operations: reported revenue subject to financial burden at rate T (interpreted as cost of complying with taxes, bribes, burdensome regulations).
  - Expected penalty of being caught, P, is increasing in α and output y, with P = C(α)y and C(.) convex.
  - Firm incomes after burden and informal income:
    - Period 1 profit: π1 = (1 − α1)(1 − T)y1 − C(α1)y1 (written in source as 11111 (1   (1)   )()Ty  Cyπαα=−−− ).
    - Period 2 profit: π2 = (1 − α2)(1 − T)y2 − C(α2)y2 + (1 − α2)Lr (written in source as 22222 (1   (1)   )()(1)    Ty   CyrLπαα=−−−−+ ).
  - Firms trade off tax/regulatory burden vs. expected penalty and reduced access to bank credit.

- Credit ceiling and determinants:
  - Firm faces credit ceiling L = L(α1; k1, q), with partial derivatives specified in source:
    - L' with respect to α: ∂L/∂α < 0 (credit ceiling decreasing in firm informality).
    - ∂L/∂k1 > 0 (credit ceiling increasing with initial capital).
    - ∂L/∂q > 0 (credit ceiling increasing with quality of legal and financial institutions).
  - Credit ceiling reflects requirement for official documentation, verifiable revenue, enforceable contracts; hiding profits increases information asymmetries and reduces lending.
  - Informal credit markets can provide financing without reliance on formal documentation, based on personal relationships and private enforcement (McMillan and Woodruff, 1999).

- Firm’s optimization and cases:
  - Firm maximizes lifetime profits subject to credit constraint; formal problem summarized in source (see Appendix 1 for details).
  - Two cases:
    - Credit ceiling not binding: L(α1; k1, q) > L (i.e., constraint not binding).
      - Then firm’s share of informal operations is decreasing in tax and regulatory burden: dαi/dT = 1/C′′(αi) > 0 (i = 1,2).
      - Larger tax and regulatory burden discourages bank financing.
    - Credit ceiling binding: L(α1; k1, q) = L.

- Proposition 1 (as stated in source):
  - A larger tax burden discourages bank financing when the disincentive for investment and the sensitivity of α with respect to the tax burden are sufficiently large.

### Empirical approach and data (as described)
- Data:
  - Firm-level data set for 26 countries in Eastern Europe and Central Asia.
  - Motivation: financial sectors in transition countries experienced rapid credit growth and diversification (EBRD, 2006); bank loans still play limited role in enterprise financing in parts of the region; business regulation costs, weak tax administration, poor institutions, and weak property rights are major obstacles (World Bank, 2007b).

- Empirical tests and robustness:
  - Tests predictions from model: relationship between informality and bank credit and role of tax and legal institutions.
  - Examine firm reliance on bank credit, use of informal credit, and firms’ self-reported financing constraints.
  - Include interactions between firm formality and institutional measures (legal quality, tax administration).
  - Control for firm transparency and performance.
  - Address potential endogeneity using instrumental variables; instrumental-variable estimates leave results unchanged.

### Key implications and policy-relevant findings
- Institutional quality matters for access to bank finance:
  - Stronger legal environment reduces informality and encourages bank borrowing even for financially opaque informal firms.
  - Weak tax administration increases reliance on informal credit, especially for informal firms.
- Tax and regulatory burden:
  - Higher tax and regulatory burden increases informality and discourages bank financing.
  - High tax compliance costs at the country level are associated with lower firm use of bank credit.
- Financial development:
  - Underdeveloped financial systems are associated with lower firm use of bank credit.
- Policy considerations (implied by results and model):
  - Reduce tax compliance burdens and streamline regulations to lower incentives for informality.
  - Strengthen tax administration to reduce interaction effect amplifying informality’s negative impact on bank credit.
  - Improve legal and financial institutions to expand credit ceilings for firms, increase enforceability, and encourage formal-sector borrowing.
  - Encourage measures that increase firm transparency (documentation, audited statements) to facilitate access to bank finance.

*Italic source attribution: Content derived from _wp0894 - References.*

### Appendix 1 for proof).

### _wp0894 - Appendix 1 for proof)

### Theoretical framework and key propositions
- Model setup:
  - Investment demand k2 remains negatively related to the tax burden; higher tax burden reduces demand for bank financing through reduced investment.
  - When the credit ceiling constraint is binding, the model reduces to:
    - 11(;,)LL   kqα=    (5)
    - 11(,, )hT q kα=    (6)
  - Assumptions:
    - 1'(    ; .)0Lα<.
    - Differentiation of (6) implies 1/0qα∂∂<: a better quality of legal and financial environment reduces propensity to operate informally.
    - Since credit ceiling L is decreasing in 1α and 11(;,)0Lkqqα∂∂>, amount of bank financing L is increasing in quality of legal and financial institutions.
- Proposition 2:
  - A better quality of legal and financial institutions increases bank financing.

### Empirical strategy
- Reduced form model estimated:
  - iijijijijjijijij XqTInfZInffCreditεββββγβ++++++== 43210 ),( (7)
  - Dependent variable: Credit (extent firms rely on bank credit for external financing).
  - Key explanatory variables:
    - Inf: share of firm's operations conducted informally (percentage of sales not reported for tax purposes).
    - Z_i: includes tax burden (T), legal system quality (q), firm-level and country-wide controls (X_i).
  - Firm-level controls: firm age, size, industry dummies, government ownership dummy, exporter dummy, corporation dummy.
  - Country-level controls: log of real GDP per capita, CPI inflation, real GDP growth.
- Estimation:
  - Dependent variables are 0–1 indicators; Probit model with maximum likelihood and heteroskedasticity-robust standard errors.
  - Economic effects evaluated at the mean value of control variables.

### Data and variable construction
- Data source:
  - 2005 Business Environment and Enterprise Performance Survey (BEEPS) by the World Bank and the EBRD.
  - Sample: firms in 26 transition countries in Eastern Europe and Central Asia.
  - Analysis restricted to firms that have external financing sources.
- Dependent variables (binary):
  - Bank Credit: 1 if firm relies on bank credit for working capital or new investment, 0 otherwise.
  - Informal Credit: 1 if firm relies on money-lenders or family and friends for working capital or new investment, 0 otherwise.
  - Financing Obstacle: 1 if firm reports availability of financing as a “major constraint” (rating of 4 on 1–4 scale), 0 otherwise.
- Key descriptive statistics:
  - Over 56 percent of firms use bank credit.
  - 29 percent use informal credit sources.
  - 22 percent report availability of financing as a major obstacle.
  - Sales to tax authorities are not fully reported by 39 percent of firms; average level of informality is 11 percent.
  - Sample averages and ranges (selected):
    - Bank Credit: Obs 4226, Mean 0.56, Std. Dev. 0.50, Min 0, Max 1
    - Informal Credit: Obs 4226, Mean 0.29, Std. Dev. 0.45, Min 0, Max 1
    - Financing Obstacle: Obs 4085, Mean 0.22, Std. Dev. 0.41, Min 0, Max 1
    - Informality: Obs 4226, Mean 11.20, Std. Dev. 19.24, Min 0, Max 98
    - Index of legal quality: Obs 3626, Mean 0.91, Std. Dev. 0.63, Min 0, Max 2
    - Private Credit (percent of GDP): Obs 4226, Mean 29.64, Std. Dev. 15.39, Min 6.19, Max 59
    - Time to pay taxes (hours per year): Obs 4226, Mean 486.82, Std. Dev. 547.82, Min 9, Max 2185
    - GDP per capita (log): Obs 4226, Mean 7.82, Std. Dev. 0.86, Min 5.47, Max 9.34
    - Inflation (annual percent): Obs 4226, Mean 5.84, Std. Dev. 4.31, Min 0.51, Max 7.30
    - Real GDP growth (percent): Obs 4226, Mean 4.99, Std. Dev. 2.81, Min 0.11, Max 4.55
    - Rule of law: Obs 4226, Mean -0.17, Std. Dev. 0.71, Min -1.32, Max 1

### Main empirical findings
- Baseline results (Table 3):
  - Informality is statistically significantly and negatively associated with reliance on bank financing:
    - Marginal effect: probability of relying on bank credit increases by 2 percentage points for every 10 percent reduction in informality at the sample average of 11 percent.
  - Informality is positively and significantly associated with reliance on informal credit.
  - Informality associated with higher incidence of severe financing constraints (higher probability of rating availability of finance as a major obstacle).
  - Firm size effects:
    - Probability of relying on bank credit decreases by 27 percent for small firms compared to large firms (marginal effect reported).
    - Reliance on informal sources increases by 13 percent for small firms.
- Business environment effects (Table 4):
  - Weaknesses in tax administration negatively and significantly affect reliance on bank credit.
    - Marginal example: firm reporting frequent bribes to deal with taxes has a 5 percent lower probability of using bank credit (column 4 marginal effect).
  - Legal quality positively associated with reliance on bank credit.
    - Improvement in legal quality index from sample average 0.91 associated with a 4-percent increase in probability of using bank credit (estimate implication).
  - Tax administration weaknesses positively associated with use of informal credit and higher probability of reporting financing as a major obstacle:
    - A firm that sees tax administration as a major obstacle has a 6 percent higher probability of using informal credit and a 12 percent higher probability of reporting being financially constrained (columns 8 and 12 estimates).
- Formal vs. informal firms (Table 5):
  - Informal firms benefit more from improvements in legal system quality: an improvement in legal quality index results in a 7 percent increase in probability of informal firms using bank credit.
  - Frequency of bribes reduces bank credit use by formal firms; tax administration constraints reduce bank credit use by informal firms.
  - Better legal environment lowers reliance on informal credit by formal firms; tax/regulatory constraints increase use of informal credit by informal firms (e.g., weaknesses in tax administration result in an 8 percent increase in probability informal firms rely on informal financing and a 4 percent increase for formal firms).
- Country-level institutional effects (Table 6):
  - More developed financial systems (private credit/GDP) and better rule of law associated with higher use of bank credit.
  - Higher tax compliance costs (time to pay taxes) negatively associated with reliance on bank credit.
  - Example quantification (Ukraine vs. Estonia):
    - If Ukraine took the same number of hours per year to comply with taxes as Estonia (104) instead of Ukraine’s level (2185), there would be a 4-percentage point increase in the probability that Ukrainian firms rely on bank credit (using column 4 results).
  - Firms in richer countries with lower inflation rely more on bank credit; GDP growth is insignificant in bank credit regressions.
  - Poorer and faster growing countries tend to rely more on informal financing.
- Interaction effects (Table 7):
  - Informal firms’ reliance on bank credit is most severely affected by weaknesses in the tax system.
  - Interaction coefficients of informality with financial sector development and rule of law are insignificant.
  - Supports model prediction that where effective tax burden is high, informal firms have greater incentive to forego bank credit.
- Firm transparency and performance controls (Tables 8–9):
  - External auditing is positively and significantly associated with reliance on bank credit.
  - Use of international accounting standards (IAS) not significantly associated with bank credit in baseline; however interaction results show that the negative relationship between informality and bank credit is less likely when firm uses international accounting standards.
  - Past sales growth is a significant predictor of bank financing.
  - Larger fixed assets increase probability of bank finance and reduce reliance on informal credit.
  - Results for informality remain robust to inclusion of auditing, IAS, sales growth, and assets.
- Instrumental variables results (Table 10):
  - Instrument: average informality in firms within the same location (correlation with firm-level informality = 0.42).
  - First-stage F-test indicates the instrument is strong enough.
  - IV estimates: instrumented informality remains negative and significant for use of bank credit and perceived financing obstacles; positive and significant for use of informal credit.
  - Reported IV coefficients (selected):
    - Informality coefficient in Bank Credit IV regression: -0.005 (standard error 0.003, significant at 10 percent in column shown).
    - Average Informality first-stage coefficient: 1.039 (standard error 0.038, significant at 1 percent).

### Policy implications and conclusions
- Empirical evidence:
  - Informality is robustly and significantly associated with lower access to and use of bank credit and higher reliance on informal financing.
  - Firm reliance on bank credit is positively associated with legal environment quality and negatively associated with weaknesses in tax administration.
  - The negative relationship between informality and access to credit is stronger in countries with weak tax administrations and high tax compliance costs.
- Policy conclusions:
  - Policies that reduce tax and regulatory constraints and improve legal environment reduce incentives for firms to operate informally by:
    - Increasing benefits of accessing formal credit markets.
    - Reducing costs of participating in formal financial systems.
  - Importance of improving tax policy and strengthening tax administration to reduce the size of the informal economy and encourage financial deepening.

*Source: _wp0894 - Appendix 1 for proof).*

### Appendix I: Solving the Model

### Appendix I: Solving the Model

### Firm's problem and Lagrangian
- The firm's problem is written as (equation formatting preserved from source):
  - (1)  ma x     (1   (1)   )   (   )()   (   )
    (1   (1)   )   (() )
    1
    () (() (1 )
    1
    . .  (    ;   ,   ),      '(.)    0,       ''(.)    0
    L
    Tfk   Cfk
    TfL
    CfLrL
    r
    st  L   Lk  qLL
    αα
    αα
    απα
    απα
    α
    −−−
    −−+
    ⎛⎞
    +
    ⎜⎟
    −+−+
    +
    ⎝⎠
    ≤<<

- The Lagrangian is given by:
  - () 1111 211 11 211  (1   (1)   )   (   )()   (   )
    (1   (1)   )   (() )
    1
    (( ; ,)   )
    ()() (1 )
    1
    LTfkCfk
    TfL
    LkqL
    CfLrL
    r
    αα
    απα
    γα
    απα
    =−−−
    −−+
    ⎛⎞
    ++−
    ⎜⎟
    −+−+
    +
    ⎝⎠

### First-order conditions (FOC)
- FOC with respect to 1α (equation A1):
  - ()() 11222211   (   )'()    1(1   (1)   )    '(    )()    '(    )(1)'(    ;    ,   )0fk  T  CrT f kCf kr  Lk qαααγα−++−−  −   ++=  (A1)
- FOC with respect to 2α (equation A2):
  - 2'()0 TCα−=                                                           (A2)
  - Thus 2α is a function of T, 2()gTα=.
- FOC with respect to L (equation A3):
  - 2222  (1   (1)   )    '(    )()    '(    )(1)0   Tf k    Cf k    rrααγ−−−−−+ =                             (A3)

### Case 1: Credit constraint ceiling not binding (11(;,)LL   kqα<)
- When the credit limit constraint is not binding, the Lagrangian multiplier γ = 0.
- Combining equations (1) and (2) and solving (A1)-(A3) yields 4 equations in 4 unknowns (1 2 2 , , , k L α α):
  - '(    )0 i TCα−=, for 1, 2i=                        (A4)
  - () 222  (1   (1)   )()'(    )0TCfk rαα−−   −−=                 (A5)
  - 21111 (1   (1)   )   (   )()   (   )0kTfkCfkLαα−−−+−=                               (A6)
- Determination:
  - 1α and 2α are determined by (A4).
  - 2k and L are jointly determined by (A5) and (A6).
- Envelope theorem derivatives with respect to T:
  - 22 (1)    '(    )0fkα−−< 
  - 11 (1)   (   )0fkα−> 
- Derivatives of 2k with respect to T:
  - 31 222 222 (1)    '(    ) (1   (1)() )    ''(    ) kfk TTCfk α α α ∂− = ∂−−− ,
  - 2 11 (1)   (   ) k αfk T ∂ =− − ∂ .
- Condition for L to fall as T increases:
  - 22 11 222 (1)    '(    ) (1)   (   ) (1   (1)() )    ''(    ) fk fk TCfk α α αα − − > − −−  −
  - Interpretation: L falls with T when the curvature of the production function is sufficiently small.

### Case 2: Credit constraint ceiling binding (11(;,)LL   kqα=)
- When the credit ceiling binds, substitute out γ by combining (A-1) and (A-3) to obtain (A7):
  - ()() ()()() 21111 1111 '(    )    1   (1(   ))(   (   ))'(    ;    ,   )(   )'(    ) ()'( )1'( ;,)() 0 f  kgT  T  C gTLk qf k   T  C fk  T  Cr   Lk q r αα αα −−−+− +− +−=
    (A7)
- From equation (1) we have (A8):
  - () 211111 (;,) 1(1   )() ()kL kqTC    fkααα=+−−− (A8)
- Solving for 1α by substituting 2k out from (A7) and (A8) yields:
  - 1α as a function 11 (,, )hT q kα=.
- Derivation of L with respect to q when 11(;,)LL   kqα=:
  - Define F1 and F2 as:
    - ()() ()()() 121111 1111 '(    )    1   (1(   ))(   (   ))'(    ;    ,   )(   )'(    ) (   )'(    )   1'(    ;    ,   )(  )0 FfkgTTCgT L  kqfkTC fk  T  Cr   Lk q r αα αα =−−−+− +− +−=,
    - () 2211111 (    ;    ,   )(1   (1)   )()(   )FkL  kqTCfkααα=−−−−  −,
  - The Jacobian determinant (note 23) would be positive and the determinant of the 2x2 matrix with elements 12 Fk∂∂, 1 Fq∂∂, 22 Fk∂∂, and 2 Fq∂∂ would be negative.
  - This implies ∂1α/∂q is negative.
  - Since L decreases with 1α and 11(;,)0Lkqqα∂∂>, it follows that L increases with q.

- Additional sign conditions and notes (from footnote 24):
  - γ is positive when the credit ceiling constraint is binding.
  - By assumption, 0),;(' 11 <qkLα and 11 '(    ;    ,   )0Lkqqα∂∂<.
  - Then, by (A1) 0)(' 1 <−αCT because the second and third terms in (A1) are positive.
  - By (A-3) 2222  (1    (1)   )    '(    )()    '(    )0Tf k    Cf k    rαα−−−−> ;
  - By (A7) 0))(')((),;(' 1111 <−+ααCTkfqkL.

---

### Appendix II: Countries in the Sample (summary)
- Country list with Number of Firms, Percent, Cumulative (as presented in source):
  - 1Albania862.042.04
  - 2Armenia2044.836.86
  - 3Azerbaijan, Rep. of511.218.07
  - 4Belarus1523.611.67
  - 5Bosnia & Herzegovina1062.5114.17
  - 6Bulgaria1724.0718.24
  - 7Croatia1393.2921.53
  - 8Czech Republic2084.9226.46
  - 9Estonia1022.4128.87
  - 10Georgia571.3530.22
  - 11Hungary3919.2539.47
  - 12Kazakhstan2245.344.77
  - 13Kyrgyz Republic852.0146.78
  - 14Latvia1473.4850.26
  - 15Lithuania1232.9153.17
  - 16Macedonia, FYR1413.3456.51
  - 17Moldova1533.6260.13
  - 18Poland4651171.13
  - 19Romania3117.3678.49
  - 20Russia2145.0683.55
  - 21Serbia & Montenegro751.7785.33
  - 22Slovak Republic1152.7288.05
  - 23Slovenia1262.9891.03
  - 24Tajikistan521.2392.26
  - 25Ukraine2806.6398.89
  - 26Uzbekistan471.11100
- Total: 4,226 100

---

### Appendix III: Variables and Sources (selected substantive definitions)
- Bank Credit
  - Definition: Dummy variable that takes on the value 1 if the firm relies on external sources (local and foreign banks) to finance either working capital or new investment, zero otherwise.
  - Original Source: Business Environment and Enterprise Performance Survey (BEEPS, 2005)
- Informal Credit
  - Definition: Dummy variable that takes on the value 1 if the firm relies on informal sources (money-lenders or family and friends) to finance either working capital or new investment, zero otherwise.
  - Source: BEEPS, 2005
- Financing obstacle
  - Definition: Dummy variable constructed on the basis of the question: how problematic is access to financing (e.g., collateral required or financing not available) for the operation and growth of your business: no obstacle (1), a minor obstacle (2), a moderate obstacle (3), or a major obstacle (4)?
  - Source: BEEPS, 2005
- Informality (percentage of sales not declared to tax authorities)
  - Definition: What percentage of total annual sales would you estimate the typical firm in your area of business reports for tax purposes?
  - Source: BEEPS, 2005
- Tax administration constraint
  - Definition: Dummy variable constructed on the basis of the question: how problematic is tax administration for the operation and growth of your business: no obstacle (1), a minor obstacle (2), a moderate obstacle (3), or a major obstacle (4)?
  - Source: BEEPS, 2005
- Bribes - taxes and tax collection
  - Definition: Dummy variable constructed on the basis of the question: how often would a firm like yours make unofficial payments/gifts to deal with taxes and tax collection in a given year: never(1), seldom(2), sometimes(3), frequently(4), usually (5), always(6)?
  - Source: BEEPS, 2005
- Legal quality index
  - Definition: Additive index based on questions: how often do you associate the description “fair and impartial”, “honest”, and “able to enforce its decisions” with the court system in resolving business disputes: always (1), usually (2), frequently (3), sometimes (4), seldom (5), never (6)?
  - Source: BEEPS, 2005
- Government ownership
  - Definition: Dummy variable that takes on the value 1 if any government agency or state body has a financial stake in the ownership of the firm, zero otherwise.
  - Source: BEEPS, 2005
- Corporation
  - Definition: Dummy variable that takes on the value 1 if the firm is organized as a corporation, and zero if the firm is organized as a cooperative, sole proprietorship, partnership, or some other legal form.
  - Source: BEEPS, 2005
- Exporter
  - Definition: Dummy variable that takes on the value 1 if firm exports, zero otherwise.
  - Source: BEEPS, 2005
- Age
  - Definition: Business age calculated as the difference between the year of the survey minus the year when the firm started operations.
  - Source: BEEPS, 2005
- Firm-size dummies
  - Definition: A firm is defined as small if it has between 2 and 49 employees, medium if it has between 50 and 249 employees, and large if it has more than 250 employees.
  - Source: BEEPS, 2005
- Audit
  - Definition: Dummy variable that takes on the value 1 if the firm has an external auditor, zero otherwise.
  - Source: BEEPS, 2005
- Account
  - Definition: Dummy variable that takes on the value 1 if the firm uses international accounting standards, zero otherwise.
  - Source: BEEPS, 2005
- Assets
  - Definition: Logarithm of the book value of the physical production assets owned and used by the firm (land, buildings, equipment).
  - Source: BEEPS, 2005
- Sales growth
  - Definition: Estimate of the firm’s sales growth over the past three years.
  - Source: BEEPS, 2005
- Private credit
  - Definition: Private credit by deposit money banks over GDP.
  - Source: Beck, Demirgüç-Kunt, and Levine(2000), updated (2007)
- Rule of law
  - Definition: Principal component indicator of survey indicators measuring the quality of contract enforcement, the police and the courts, as well as the likelihood of crime and violence.
  - Source: Kaufmann, Kraay, and Mastruzzi (2006)
- Time to pay taxes
  - Definition: Time taken to prepare, file and pay (or withhold) three major types of taxes: corporate income tax, value added or sales tax and labor tax, including payroll and social security contributions.
  - Source: World Bank Doing Business Survey (2005)
- GDP per capita
  - Definition: Log of per capita GDP in constant 2000 U.S. dollars.
  - Source: World Development Indicators (WDI)
- GDP growth
  - Definition: Growth rate of real GDP, average 2002-2005.
  - Source: WDI
- Inflation
  - Definition: Log difference of Consumer Price Index.
  - Source: IFS, line 64

*Source: _wp0894 - Appendix I: Solving the Model*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp0894.pdf_