Credit and saving constraints in general equilibrium: Evidence from survey data

Granda Carvajal, Catalina; Hamann S., Franz A.; Tamayo, Cesar E. Working Paper Credit and saving constraints in general equilibrium: Evidence from survey data IDB Working Paper Series, No. IDB-WP-808 Provided in Cooperation with: Inter-American Development Bank (IDB), Washington, DC Suggested Citation: Granda Carvajal, Catalina; Hamann S., Franz A.; Tamayo, Cesar E. (2017) : Credit and saving constraints in general equilibrium: Evidence from survey data, IDB Working Paper Series, No. IDB-WP-808, Inter-American Development Bank (IDB), Washington, DC, https://hdl.handle.net/11319/8283 This Version is available at: https://hdl.handle.net/10419/173872 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by-nc-nd/3.0/igo/legalcode IDB WORKING PAPER SERIES Nº IDB-WP-808 Credit and Saving Constraints in General Equilibrium Evidence from Survey Data Catalina Granda Franz Hamann Cesar E. Tamayo Inter-American Development Bank Institutions for Development Sector May 2017 May 2017 Credit and Saving Constraints in General Equilibrium Evidence from Survey Data Catalina Granda Franz Hamann Cesar E. Tamayo Contact: Cesar Tamayo, [email protected]. Copyright © Inter-American Development Bank. This work is licensed under a Creative Commons IGO 3.0 Attribution-NonCommercial-NoDerivatives (CC-IGO BY-NC-ND 3.0 IGO) license (http://creativecommons.org/ licenses/by-nc-nd/3.0/igo/legalcode) and may be reproduced with attribution to the IDB and for any noncommercial purpose, as provided below. No derivative work is allowed. Any dispute related to the use of the works of the IDB that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the IDB's name for any purpose other than for attribution, and the use of IDB's logo shall be subject to a separate written license agreement between the IDB and the user and is not authorized as part of this CC-IGO license. Following a peer review process, and with previous written consent by the Inter-American Development Bank (IDB), a revised version of this work may also be reproduced in any academic journal, including those indexed by the American Economic Association's EconLit, provided that the IDB is credited and that the author(s) receive no income from the publication. Therefore, the restriction to receive income from such publication shall only extend to the publication's author(s). With regard to such restriction, in case of any inconsistency between the Creative Commons IGO 3.0 Attribution-NonCommercial-NoDerivatives license and these statements, the latter shall prevail. Note that link provided above includes additional terms and conditions of the license. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the Inter-American Development Bank, its Board of Directors, or the countries they represent. Cataloging-in-Publication data provided by the Inter-American Development Bank Felipe Herrera Library Granda, Catalina. Credit and saving constraints in general equilibrium: evidence from survey data / Catalina Granda, Franz Hamann, and César E. Tamayo. p. cm. — (IDB Working Paper Series ; 808) Includes bibliographic references. 1. Credit-Colombia-Econometric models. 2. Saving and investment-Colombia- Econometric models. 3. Financial institutions-Colombia. I. Hamann, Franz. II. Tamayo, César E. III. Inter-American Development Bank. Connectivity, Markets and Finance Division. IV. Title. V. Series. IDB-WP-808 http://www.iadb.org 2017 Credit and Saving Constraints in General Equilibrium: Evidence from Survey Data∗ Catalina Granda†Franz Hamann‡Cesar E. Tamayo§ May 3, 2017 Abstract In this paper, we build a heterogeneous agents-dynamic general equilibrium model wherein saving constraints interact with credit constraints. Saving constraints in the form of fixed costs to use the financial system lead households to seek informal saving instruments (cash) and result in lower aggregate saving. Credit constraints induce misallocation of capital across producers that in turn lowers output, productivity, and the return to formal financial instruments. We calibrate the model using survey data from a developing country where informal saving and credit constraints are pervasive. Our quantitative results suggest that completely removing saving and credit constraints can have large effects on saving rates, output, TFP, and welfare. Moreover, we note that a sizable fraction of these gains can be more easily attained by a mix of moderate reforms that lower both types of frictions than by a strong reform on either front. Keywords: saving constraints, credit constraints, financial inclusion, misallocation, saving, formal and informal financial markets. JEL Classification Numbers: E21, E44, G21, O11, O16 ∗Amalia Rodríguez provided valuable assistance in processing the microdata from the ELCA survey. We thank seminar participants at ICESI, Eafit, Banco de la República and Universidad del Rosario for comments and suggestions. All errors and omissions are our own. The views expressed in this paper are entirely those of the authors and no endorsement by the Banco de la República, the Inter-American Development Bank, their Board of Directors, or the countries they represent is expressed or implied. †Universidad de Antioquia, [email protected] ‡Banco de la República, [email protected] §Inter-American Development Bank, [email protected] 1 1 Introduction Financial inclusion –broadly defined– has become a priority for economists and policy makers trying to advance the development agenda.1Indeed, in recent years the longstanding goal of improving access to credit in developing countries has been joined by a growing interest in the role that saving should have in a comprehensive financial inclusion strategy. While the literature on credit frictions is well developed and includes both empirical and theoretical contributions, the literature on the causes and consequences of exclusion from formal (i.e., through financial institutions) saving markets mostly comprises field experiments in relatively small communities.2In fact, little is known about the general equilibrium effects of saving constraints or the way in which they may interact with other frictions, such as those found in credit markets. Our goal in this paper is to present a framework that can be used to quantify these effects and study these interactions. More specifically, we develop a model of heterogeneous agents in which financial market frictions distort credit and saving decisions by households and firms. Households save for precautionary reasons using either a deposit contract with a bank (formal saving) or cash (informal saving). Saving constraints result from the fact that using the deposit contract is costly. Entrepreneurs can access credit markets when deciding on their capital input, but face collateral requirements due to limited enforcement problems. Saving constraints lead households to seek informal saving instruments (cash) and result in lower aggregate saving. Credit constraints induce misallocation of capital across producers, and in turn this lowers output, productivity, and the return to formal financial instruments. To discipline the model parameters, we use data from the Colombian Longitudinal Survey, which contains income and occupational data as well as detailed information concerning financial decisions by households. In particular, we use data from this household survey to obtain saving rates and observe the incidence of informal saving. We complement these figures with data from credit markets, firm dynamics, and macroeconomic aggregates. Our counterfactual exercises suggest that the production efficiency and welfare losses from financial market distortions can be substantial. In fact, completely eliminating saving and credit constraints can double the fraction of households who save and increase the average saving rate by over 60%. Moreover, it can increase total factor productivity (TFP) by 5.6%, output by 25%, and households welfare by 50%. While efficiency gains (TFP and output) are mainly due to 1According to the Alliance for Financial Inclusion (2016), by 2015 over 35 countries had committed to implementing or had already implemented financial inclusion strategies. 2Reference studies from the credit frictions literature are Kaplan & Zingales (1997) and Buera et al. (2011), and from the saving constraints literature are Dupas & Robinson (2013) and Karlan et al. (2014). 2 better capital allocation, the welfare effects result from both higher average consumption and smoother consumption profiles made possible by the first-best saving policy. Interestingly, a sizable fraction of these gains can be attained by a mix of moderate policy reforms that address distortions in both credit and saving decisions. This paper is related to a number of recent studies addressing the interaction between formal and informal financial markets in developing countries. In this respect, Wang (2014) develops and estimates a dynamic equilibrium model of borrowing and saving decisions that allows him to interpret Thailand’s financial reform in 2001 as one that reduced formal borrowing interest rates, lowered costs of access to credit, and relaxed collateral constraints. This reform in turn led to an increase in the proportion of households borrowing formally and to a fall in informal interest rates. He finds that the welfare gains from these policies are smaller than suggested by previous studies that disregard informal saving options. Furthermore, two streams of the financial development literature are relevant for the present study.3One stream has been looking into the determinants of access to and use of savings instruments and their effect on economic outcomes in recent years. This strand of the literature has been mainly focused on the extensive margin, and includes both cross-country studies (Demirgüç-Kunt & Klapper,2013;Rojas-Suarez & Amado,2014) and field experiments inside villages or larger regions within a country (see Dupas & Robinson,2013;Kast & Pomeranz, 2014;Prina,2015, to name a few). Overall, this strand of the literature shows that the world’s population –particularly in poor regions– often saves using formal or informal instruments that entail high risk, are costly, and have limited functionality. This leads to low saving rates, with significant welfare consequences: reduced consumption smoothing, low resilience to shocks, and foregone profitable investment opportunities. In a survey of this literature, Karlan et al. (2014) group constraints to saving into five categories; transaction costs, lack of trust and regulatory barriers, information and knowledge gaps, social constraints, and behavioral biases. This paper focuses on market frictions that hinder the supply of savings products. The other stream has been devoted to quantitatively examining the impact of financial frictions on economic development. Within this strand of the literature, some studies argue that imperfect contract enforceability generates distortions in the allocation of capital across production units that in turn leads to aggregate productivity losses. These studies typically characterize imperfections in financial markets in the form of collateral constraints (Buera et al.,2011;Buera & Shin,2013;Midrigan & Xu,2014). The model developed in this paper, as in these studies, 3For a recent survey of the financial development literature, see Fernandez & Tamayo (2017). 3 features financial frictions due to limited enforcement and allows studying their implications for capital misallocation. In a similar vein, several studies focus on the effect of credit market imperfections on occupational choice. Among these, Buera et al. (2011) and Buera & Shin (2013) highlight that financial frictions trigger distortions in the allocation of entrepreneurial talent that result in low aggregate productivity and output. Somewhat similarly, Antunes et al. (2008) show that intermediation costs and contract enforcement may explain cross-country differences in entrepreneurship and other development indicators. Moreover, Erosa (2001) finds that costly intermediation, when individuals choose occupations, has nontrivial consequences for saving behavior and production efficiency. Relying on this strand of the literature, recent studies have attempted to evaluate the impact of relaxing constraints to financial deepening and inclusion on growth and income inequality. In this regard, Dabla-Norris et al. (2015) analyze three types of financial frictions: participation costs, collateral requirements, and costly monitoring. Their results suggest that the effect of policies alleviating these frictions individually or jointly depends on country-specific characteristics. For Colombia, using the Dabla-Norris et al. (2015) approach, Karpowicz (2014) finds that lowering collateral requirements promises higher growth while inequality would be better tackled through reductions in participation costs. Missing in the literature are studies quantifying the efficiency gains from ameliorating distortions in the allocation of credit and savings through formal financial instruments. Filling this void is important for at least two reasons. First, development experiments have revealed that general equilibrium effects from relocating savings toward formal financial instruments are important (Flory,2017). Second, a significant determinant of the demand for formal savings instruments is its return, which is an endogenous outcome of the financial intermediation process thus affected by credit allocation. The paper is organized as follows. In Section 2, we present some empirical regularities pertaining to barriers to financial inclusion and patterns of saving behavior in Colombia. The main aspects of the model economy are described in Section 3. Simulations and policy scenarios are presented in Section 4. Section 5concludes. 2 Empirical Regularities Increasing awareness of the role of the financial sector and the importance of financial inclusion have led to the construction of a growing number of databases and surveys, providing evidence on 4 access to and use of financial services by households and firms in developing countries. With the aim to lay the ground for our theoretical model, we use some of these data sources to build a set of empirical regularities regarding saving behavior in Colombia. The following lines reflect upon this exercise, while featuring some comparisons with other Latin American peers and countries with similar levels of development. According to the Global Financial Inclusion Indicators (Global Findex), in Colombia about 72% of the respondents who were savers in 2014 reported that they save outside the financial system. This percentage places the country in the 60th position among 159 nations in terms of use of informal saving instruments such as cash or saving chains. Further, given its level of development (proxied by PPP GDP per capita), Colombia appears to be somewhat of an outlier in this respect (see Figure 1, left panel). Moreover, Colombia is one of the countries where most respondents (about 20%) did not have an account at a financial institution because they found it too expensive.4In this sense, Colombia holds the 37th position among 142 countries, above some Latin American peers such as Brazil, Chile, Uruguay, and Costa Rica. As with informal savings, Colombia’s percentage of financially excluded for cost-related reasons is higher than would be expected given its income level (Figure 1, right panel). Figure 1: Informal savings and the importance of cost in access to bank accounts 0 .2 .4 .6 .8 1 % respondents who saved outside the financial system 4 6 8 10 12 GDP per capita (log) COL R2=0.6382 COL R2 = 0.4257 0 .1 .2 .3 .4 .5 % respondents without a bank account due to cost 4 6 8 10 12 GDP per capita (log) Source: Global Findex 2014. In order to look at these issues in more detail, we use the financial chapter of the Colombian Longitudinal Survey (henceforth ELCA), which provides specific information on financial 4The Global Findex survey asks respondents if they own an account at a “bank or credit union” (or another financial institution, where applicable, like a cooperative in Latin America) (Demirgüç-Kunt & Klapper,2013, p. 313). 5 places Colombia as the fourth country in Latin America in which firms find themselves more financially constrained (see Figure 6, left panel). Moreover, Figure 6(right panel) suggests that there is a positive relation between the tightness of financial constraints and the likelihood of saving outside the financial system, presumably because of the low returns channel described above. 3 A Model of Credit and Saving Constraints In this section, we develop a dynamic general equilibrium model with saving constraints at the household level and wherein entrepreneurs are credit constrained. Some of the key features of the model come directly from the evidence found in the ELCA and the World Bank databases: saving in banks is costly, people save mainly for consumption smoothing, the main alternative to bank saving is cash, and firms face borrowing constraints. The economy is populated by a measure Nof workers and a unit measure of entrepreneurs. Both workers and entrepreneurs are heterogeneous with respect to their productivity and seek to maximize lifetime utility given by max(E0 ∞ ∑ t=0 βtu(Ct)),(2) where period utility is of the constant relative risk aversion form: u(C) = C1−χ 1−χ,(3) with χ>0. Entrepreneurs can borrow and save with financial intermediaries, but face a collateral requirement that constrains the amount they can borrow. Workers face uninsurable idiosyncratic labor income risk and have access to financial markets. There are two types of financial instruments available: a one-period risk-free asset (formal) and cash (informal). Entrepreneurs. Entrepreneurs have access to a decreasing returns technology that uses labor Land capital Kto produce a consumption good Y. Specifically: Yt=At[aexp(zt)]1−θ(Kλ tL1−λ t)θ.(4) Here, ais a permanent ability component, zis a transitory productivity component, and Ais an aggregate efficiency component. Also, fraction 0 <λ<1 of output corresponds to capital and 12 fraction 1 −λto labor, and θ<1 is the degree of decreasing returns to variable inputs. Capital depreciates between periods at rate δ. An entrepreneur’s idiosyncratic productivity consists of a permanent component aand a transitory component z. Permanent entrepreneurial ability (or talent) ais drawn at birth from a distribution Γ(·). In each period, a fraction 1 −ηof entrepreneurs die and is replaced by new ones, in which case the firms they own exit at zero market value.10 Aggregate efficiencyAt grows deterministically at a constant rate g,At=At−1g, while the transitory productivity component ztevolves over time according to a finite-state Markov process with transition probabilities π(z0,z) = Pr(zt+1|zt)and ergodic distribution ξ(z). Because of growth in A, most aggregates in this economy are non-stationary with a deterministic trend. Normalizing A0=1 and defining γ=g(1/(1−α)), such a trend can be found to be γt.11 Throughout the paper, we deal with de-trended variables only. Moreover, because ais permanent and our problem is homogeneous as in Midrigan & Xu (2014), we can scale all variables in the entrepreneur’s problem by her permanent productivity a. Henceforth, xdenotes the de-trended, scaled value of any variable X(i.e., xt≡(Xt/aγt)). Entrepreneurs decide how much to borrow (dt)and save (bt+1). Since bt=kt−dt, and b is pre-determined, choosing dtamounts to choosing kt. Further, we assume that entrepreneurs cannot fully commit to repaying loans because financial contracts are imperfectly enforceable. In particular, defaulting entrepreneurs keep a percentage 1−ϕof their capital stock; the remaining fraction ϕis recovered by the lender.12 Finally, we assume that all saving by firms is done through the one-period bank deposit and that using a bank to save requires paying a per-period fixed cost τ.13 Given prices, (r,w), an entrepreneur’s problem can be stated recursively as: V(b,z) = max b0,k,l c1−χ 1−χ+βηγ1−χ∑ z0 V(b0,z0)π(z0|z)(5) subject to c+γb0+τ= [exp(z)]1−θ(kλl1−λ)θ−(r+δ)k−wl +(1+r)b(6) 10This reflects the fact that firms exit the market for reasons not internalized by the model. It is well known that without exogenous exit some firms would eventually accumulate enough assets to overcome borrowing constraints and over time the mass of firms would grow without bound (Quadrini,2004). 11For further details, see Appendix 6.1. 12Although the collateral constraint ensures that all contracts are enforceable, if a firm were to default, it would exit the market irreversibly. 13This follows the evidence found in Didier & Schmukler (2014) that virtually all firms in Latin America own and use formal bank accounts. 13 and the collateral constraint d≤ϕk,(7) which can be rewritten as k≤b 1−ϕ.(8) Notice that our specification of the collateral requirement is virtually identical to that used in Midrigan & Xu (2014). Workers. Each worker is endowed with a unit of labor which is supplied inelastically. Labor income, however, depends upon the worker’s idiosyncratic efficiency, which is a composite of innate (permanent) ability νand transitory shocks εt. Permanent ability is drawn from a distribution Ω(·), while εtevolves over time according to a finite-state Markov process with transition probabilities ψ(ε0,ε) = Pr(εt+1|εt) and ergodic distribution µ(ε). Workers save using cash s or a one-period deposit contract with a bank q. While cash (the “informal” instrument) yields no interest, deposits (the “formal” instrument) yield a non-negative risk-free rate of return r.14 Those workers who engage in deposit saving must pay a fixed cost τfor every period they use the bank. As with the entrepreneurs, we write the workers’ problem in terms of de-trended, scaled variables (i.e., xt≡(Xt/νγt)). Given prices (r,w), a worker’s problem can be stated recursively as: W(q,s,ε) = max c,q0,s0 c1−χ 1−χ+βγ1−χ∑ ε0 W(q0,s0,ε0)ψ(ε0|ε)(9) subject to c+γq0+γs0+τIq0>0=wexp(ε)+(1+r)q+s(10) and the no-borrowing constraints q≥0,s≥0,(11) where Iq0>0is an indicator variable that equals one if the worker saves using the formal instrument, and zero otherwise. Financial intermediaries. Banks take deposits from workers and lend them to firms. Because all contracts are strictly enforceable (i.e., there is no default in equilibrium), firms pay and workers receive exactly the risk-free rate which is endogenously determined. Naturally, some 14In this sense, one might think of cash saving as a storage technology with zero returns next period. 14 firms will face a higher shadow price of capital than others depending on whether the collateral constraint binds, and some workers will face a lower return once they account for fixed costs of deposit market participation. Equilibrium. The scaled and de-trended economy has a stationary equilibrium that consists of a set of prices (w,r), stationary distributions of workers gand entrepreneurs h, and decision rules {c(q,ε),q+1(q,ε),s+1(q,ε),b+1(b,z),k(b,z),l(b,z)}, where “+1” stands for one period ahead, such that: •All workers and entrepreneurs optimize, that is, l(b,z),k(b,z),b0(b,z)solve problem (4)- (7) and c(q,ε),q0(q,ε),s0(q,ε)solve (8)-(9); •The labor market clears, ∑ b,z h(b,z)l(b,z) = N∑ ε εµ(ε); (12) and •The asset market clears, ∑ b,z h(b,z)k+1(b,z) = ∑ q,s,ε g(q,s,ε)q+1(q,s,ε)+∑ b,z h(b,z)b+1(b,z).(13) 4 Quantitative Performance In this section we describe how data are used to calibrate the model presented in Section 3. We also present a series of policy experiments that allow us to quantify the costs associated with saving and credit constraints in a developing economy such as Colombia. 4.1 Calibration The model is calibrated to be consistent with a number of features of the Colombian economy. We divide the parameter vector into two groups. The first group includes preference and technology parameters that are difficult to identify using our data (see Table 2). We assign these parameters values that are common in the existing dynamic, general equilibrium (DGE) literature. Accordingly, the period is set to one year so that the discount factor is equal to 0.958. This is a common value in studies on emerging market economies. The risk aversion coefficient is set to 2.3, which is close to the value estimated for Colombia in Prada & Rojas (2010). 15 As for the technology parameters, Zuleta et al. (2010) obtain several estimates of the factor shares that we in turn use to get a measure of returns to scale. Accordingly, the capital income share λis set equal to 0.46 and the share of variable inputs θsums to 0.85. Also, the depreciation rate δis set to 0.075 as in Hamann & Mejía (2013). Further, the survival rate of entrepreneurs η is set so that 1−η=0.07 in order to match the average firm exit rate in the manufacturing sector, as reported in Eslava et al. (2013, Table 1). Finally, the trend growth parameter γcorresponds to the long-run output growth rate and is estimated as the average annual growth rates of output from 1976 to 2012 using yearly data from the National Department of Statistics (DANE). Table 2: Preference and technology parameters Parameter Value Description Source β0.958 Discount factor DGE literature χ2.300 Risk aversion coefficient DGE literature θ0.850 Share of variable inputs Zuleta et al. (2010) λ0.460 Capital share in output Zuleta et al. (2010) δ0.075 Capital depreciation rate Hamann & Mejía (2013) 1−η0.070 Firm exit rate Eslava et al. (2013) γ1.038 Trend output growth DANE Unlike the above, parameters in the second group are chosen to replicate certain moments of the Colombian data (see Table 3). First, the transitory productivity component of workers εis assumed to evolve according to a first-order autoregressive process with Gaussian disturbances and is discretized into a 10-state Markov chain using the Rouwenhorst (1995) method. The autocorrelation coefficient ρεand the standard deviation σεare chosen to approximately match the saving rate and the fraction of non-savers obtained from the financial module of the ELCA. Similarly, the transitory productivity component of entrepreneurs zfollows an AR(1) process discretized into a 15-state Markov chain so that the persistence parameter ρzand the standard deviation σzapproximately mimic the entrepreneurial saving rate and the fraction of non-savers. Second, the permanent skill component for workers νis assumed to follow a truncated and discretized version of a Pareto distribution with probability density Ω(ν) = ων−(ω+1)for ν≥ 1. The tail parameter ωis chosen to replicate the share of labor income generated by the top 1% of workers also obtained from the ELCA. Likewise, entrepreneurial ability ais assumed to be a truncated and discretized version of a Pareto distribution with probability density Γ(a) = ζa−(ζ+1)for a≥1, so that the tail parameter ζaims to mimic the share of total income generated by the top 1% of the population computed from the National Household Survey (GEIH). Finally, we calibrate the parameters that govern the functioning of financial markets. In this 16 regard, the cost of using formal saving instruments τis set to match the fraction of savers that resort to formal financial instruments obtained from the ELCA. Also, the parameter that captures limited enforcement ϕis chosen to replicate a proxy of the credit-to-output ratio that measures the ratio of credit to enterprises (corporate plus microcredit) to private value added computed using data from Banco de la República (the Central Bank of Colombia). Table 3: Summary of calibrated parameters Param Value Description Target Source ω1.900 Tail param Pareto workers Top 1% income share (workers) ELCA ζ2.010 Tail param Pareto firms Top 1% income share (all) GEIH ρε0.675 AR(1) labor productivity % of workers who do not save ELCA σε0.235 Std dev labor productivity Workers saving rate ELCA ρz0.150 AR(1) entrep productivity % of entreps who do not save ELCA σz0.560 Std dev entrep productivity Entrepreneurs saving rate ELCA ϕ0.165 % of pledgeable collateral Credit-to-output ratio Central Bank τ0.020 Fixed cost of formal saving % of formal savers ELCA The resulting economy, as can be seen in Table 4, resembles the targeted moments fairly well. Specifically, the model economy appropriately replicates key statistics such as the workers’ saving rate, the fraction of households that saves using formal financial instruments, and the credit-to-output ratio. Also, the benchmark economy mimics the percentage of income owned by the top 1% of the workers’ and the economy-wide income distributions very closely. Yet the model is not as successful in replicating all statistics. In particular, it overpredicts the share of entrepreneurs who are savers. In this sense, it must be noted that the ELCA surveys mostly small entrepreneurs that are not representative of the full entrepreneurial population that the model aims to portray. Table 4: Calibration results Targeted moment Data Model % of workers who do not save 73.3% 62.9% % of formal savers 62.2% 63.1% Workers saving rate 12.1% 12.0% % of entrepreneurs who do not save 76.1% 20.8% Entrepreneurs saving rate 23.9% 19.4% Credit-to-output ratio 0.318 0.312 % income in top 1% (workers) 7.2% 7.1% % income in top 1% (economywide) 11.3% 11.1% 17 4.2 Counterfactual Analysis In order to study the effects of alternative financial inclusion policies, we analyze the implications and limitations of a number of policy scenarios. The first experiment aims to measure the impact of eliminating all costs associated with formal saving. This is accomplished by reducing the costs of using the financial system τfrom its calibrated value to zero. The results of such a reduction are presented in the second column of Table 5, in which, to facilitate comparison, we reproduce the performance of the benchmark economy under the label “model Colombia”. Table 5: Policy experiments Statistic Model τ= 0 τ= 0 Model “Colombia” ϕ= Colombia ϕ= Chile “Efficient” % of workers who do not save 62.9% 62.5% 33.1% 27.0% % of formal savers 63.1% 100.0% 100.0% 100.0% Workers saving rate 12.0% 11.5% 12.1% 19.2% % of entrepreneurs who do not save 20.8% 20.7% 24.8% 50.1% Entrepreneurs saving rate 19.4% 19.3% 19.6% 21.2% Credit-to-output ratio 0.31 0.32 0.72 2.35 % income in top 1% (workers) 7.1% 7.1% 7.1% 7.0% % income in top 1% (economywide) 11.1% 11.3% 10.5% 8.1% % of capital financed by firms 83.6% 83.5% 65.4% 8.9% Capital intensity (K/Y) 1.909 1.932 2.069 2.584 Aggregate output 31.540 31.860 33.324 39.504 Total factor productivity (TFP) 1.971 1.974 1.988 2.072 Net real interest rate 6.31% 4.66% 6.05% 7.59% Real wage rate 0.390 0.392 0.410 0.488 Welfare (utilitarian) Households (workers) -1,646.2 -1,613.2 -1,364.6 -657.2 Firms (entrepreneurs) -248,663.1 -250,474.5 -208,193.9 -58.5 Income distribution % income in quintile 5 59.4% 60.0% 57.0% 42.8% % income in quintiles 1 and 2 15.2% 15.1% 15.8% 22.1% % income in quintiles 3 and 4 24.4% 24.9% 27.1% 35.1% It can be seen that accomplishing τ=0 entails some benefits for households, as an important fraction of these could now save using a formal financial instrument. An important metric for comparison of policy experiments is the so-called utilitarian welfare measure. This measure 18 assigns equal weights to each household’s welfare, which in turn is calculated as the present value of intertemporal utility when every household follows her optimal plan: W∗=∑ q,s,ε W(q,s,ε)g(q,s,ε)(14) where W(q,s,ε)is as found in equation (9). Using this metric, it can be observed that eliminating the costs of formal saving increases household welfare by 2%. This effect takes place not only due to an increase in average consumption, but also because formal saving allows households to better smooth consumption in the face of income shocks. Also, it is noteworthy that a “policy” of free formal saving gives rise to an increase in aggregate output of 1% and in capital intensity of 1.2%. It is important to note that, since the economy is closed, this policy affects the interest rate to influence the observed results. However, these interest rate changes have a moderate effect on the credit-to-output ratio, which increases by only 2.2%. This occurs because the policy is not accompanied by any reforms in access to credit whatsoever; that is, there are no changes in the financial constraints faced by entrepreneurs (ϕremains constant). Indeed, the observed impact of reducing the costs of using the financial system on saving behavior is a combination of two counterveiling effects. First, a higher percentage of households save in deposits and receive returns rq, which increases their (non-labor) income. Second, a general equilibrium effect ensues, as the higher supply of loanable funds –absent any significant change in the demand for credit– lowers the interest rate (discouraging saving).15 In a small open economy, where the interest rate is more or less determined by the rate that prevails in international capital markets, the effect of reducing the costs of formal saving on welfare is much higher as the general equilibrium effect is absent. In this sense, this experiment must be considered as a lower bound for the impact that a policy incentivizing formal saving could bring about. The next experiment we consider precisely addresses the issue discussed above. That is, we complement the “formal saving policy” with a financial reform that reduces –although it does not eliminate– enforcement problems in the credit market. In particular, we raise ϕso that the resulting credit-to-output ratio increases to a level similar to the one observed in Chile, a country frequently used as a leading example for Colombia in terms of financial market development. The outcomes are presented in the third column of Table 5. A number of results from this combination of reforms are worth noticing. First, although the saving rate of savers remains roughly constant, the fraction of households that save almost 15Note the marginal fall in the saving rate and the fraction of non-savers. 19 doubles (from 37.5% to 66%), which in turn means that the aggregate saving rate increases substantially. This result is in line with the argument made recently by Inter-American Development Bank (2016, Chapter 11) that multi-faceted financial reforms are needed to promote higher domestic savings in Latin America. Given this increase in saving by workers, the fraction of firms that save and the share of capital financed by firms both fall. Capital intensity increases by 8.4%, while output rises by 5.6%. Most importantly, the welfare of both households and entrepreneurs increases substantially, by 15% and 17%, respectively. In this economy, the interest rate is higher because the productivity of capital increases, as is the wage rate. Finally, there is a moderate increase in aggregate TFP and a moderate decrease in income inequality. Figure 7ilustrates that the latter effect gives rise to a situation in which the increase in welfare that results from the combination of reforms is larger for the lowest percentiles of the income distribution. Figure 7: Financial reforms and welfare The final experiment that we consider implies relaxing completely both saving and credit constraints. This is easily done by setting τ=0 and ϕ=1. The results from this experiment are presented in the last column of Table 5under the “efficient” label. In this economy, over 70% of workers save and thus they finance virtually the entire capital stock. Most importantly, losses due to misallocation are eliminated as entrepreneurial talent becomes the only determinant of capital input (i.e., credit frictions do not constrain firm size). This point is ilustrated by Figure 8, which plots the capital input for each level of entrepreneurial ability under both parameteriza- 20 tions (“Colombia” and “efficient”). Compared with the model calibrated to Colombia, workers’ saving rate is almost 60% higher, the capital-to-output ratio is 35% higher, output is 25% higher, and TFP is over 5% higher. Welfare increases by over 50% for workers and nearly 100% for entrepreneurs. Figure 8: Entrepreneurial ability and capital allocation 5 Concluding Remarks In this paper, we used recently collected survey data to study the costs associated with saving and credit constraints through the lens of an otherwise standard heterogeneous agents setting. In our model, the costs of using financial instruments distort saving decisions by households, leading to volatile consumption profiles. These constraints interact with credit frictions to generate a vicious circle of informal savings, capital misallocation, and low returns to formal saving instruments. Our quantitative results point to potentially large gains to be made in terms of production efficiency and welfare by removing these constraints. These provide support to the importance of comprehensive strategies to develop financial markets, especially in developing countries. At the same time, our results suggest that this type of study could greatly complement the growing literature on small-scale field experiments associated with financial inclusion policies. 21