Do Public Banks in Brazil Hurt Productivity?

May 1, 2015 3:45 PM

Much is said about lagging productivity growth in Latin America, particularly in comparison to emerging markets in other regions. The Inter-American Development Bank (IADB) has just released a book entitled Rethinking Productive Development, and one need not go beyond its initial pages to find the diagnosis: Reversing pervasive low productivity in Latin America requires concerted policy efforts to restore the region's growth potential. Jose De Gregorio proposes a similar solution in From Rapid Recovery to Slowdown: Why Recent Economic Growth in Latin America Has Been Slow. Both studies argue that productivity is not manna from heaven: We must build the conditions for economies to become more productive.

Who bears this responsibility? The IADB study argues for building institutions whose chief mandate is to promote productivity, underscoring that governments should play a prominent role in this process. However, a quick bird's eye view over the region suggests that Latin America has public development institutions to spare—Brazil has its state development bank BNDES, the largest in the region; Andean countries have the CAF (Corporación Andina de Fomento), and so on.1

What appears to be lacking in Latin America is a clear vision about the role these institutions should play. Should they compete with private markets or encourage mechanisms that promote economic development? The answer to this question seems obvious, but in practice development banks and other institutions in Latin America tend to replace markets rather than address collective action failures that lead to market incompleteness. When public institutions address agency failures rather than collective action failures, the end result might be lower rather than higher efficiency. Put differently, the economy becomes less productive.

A quick growth accounting exercise for Brazil shows that productivity growth is indeed dismal (table 1).2

Table 1 Brazil: Growth accounting (percent)
Year Use of Disposable Capital Education Demographic Factors Labor market Dynamics Total factor productivity GDP Growth
2003 0.85 0.76 1.31 1.90 -3.59 1.22
2004 1.30 0.60 1.26 0.76 1.75 5.66
2005 0.89 0.58 1.22 0.58 -0.13 3.15
2006 0.53 0.90 1.19 0.23 1.14 4.00
2007 1.49 0.85 1.15 0.57 1.95 6.01
2008 1.32 0.80 1.12 1.22 0.57 5.02
2009 -1.59 0.77 1.07 -0.51 0.02 -0.24
2010 2.46 0.62 1.03 1.54 1.93 7.57
2011 0.49 0.42 0.98 0.56 1.46 3.92
2012 0.55 0.80 0.94 0.67 -1.20 1.76
2013 0.78 0.73 0.90 -0.23 0.56 2.74
2014 0.33 0.70 0.86 -0.92 -0.83 0.14

Source: IPEADATA; Bolle, M. B., and P. H. Simões, 2015 (forthcoming), "Another Lost Decade? Productivity and Potential Growth in Brazil."

How do public banks potentially hurt productivity growth in Brazil?

Public banks such as BNDES play an important role in Brazil. The inexistence of long-term credit markets creates a natural niche for the state-owned public bank with clear benefits for the economy. Moreover, during the 2008 financial crisis, BNDES' risk absorption capabilities made it an invaluable policy instrument to countervail the fallout from global turmoil. However, the expansion of the bank's balance sheet in recent years and its evolving operations call into question the optimal balance of costs and benefits associated with the institution.

Consider its current role. The public bank alone accounts for about 30 percent of the domestic credit market. Moreover, according to a recent report from the International Monetary Fund, 41 percent of BNDES' portfolio caters to the ten largest Brazilian firms, those with the highest credit ratings and consequently the least likely to be liquidity-constrained. This has some important implications: Not only is BNDES directing resources to those who have access to private capital markets, thereby displacing lending to other firms such as small and medium enterprises (SMEs), but it is doing so with substantial cost to the Treasury—an unequivocal example of resource misallocation. A large part of BNDES' funding comes from the Brazilian Treasury—to fund BNDES, the Treasury borrows from the market at the short-term nominal rate (the Selic, currently at 13.25 percent), and on-lends these resources to the public bank at Brazil's long-term interest rate, the TJLP. The TJLP embeds a considerable subsidy, as explained in a previous blog post. All loans from BNDES are linked to the TJLP. Hence, unconstrained firms have the additional luxury of tapping into subsidized credit lines, crowding out SMEs and other companies.

Evidently, this creates a number of distortions in capital markets, both macroeconomic and microeconomic in nature. Distortions misallocate resources, reduce efficiency, and hurt productivity.3 On the macroeconomic side, monetary policy is unable to influence a significant part of the demand for credit, choking off an important transmission channel and leading to higher short-term interest rates than would otherwise be warranted—this effect is well-known and much discussed. Less attention is given to microeconomic distortions associated with BNDES' operations: When BNDES lends to companies with lower risk profiles, it creates an adverse selection problem in credit markets. The firms that are left out tend to be the higher risk borrowers, those that will naturally be charged higher lending rates by private banks. Additionally, if private banks become increasingly exposed to higher risk borrowers as a result of adverse selection associated with BNDES lending, they may be subject to higher reserve requirements that act as a tax on financial intermediation. Brazil's reserve requirements are among the highest in the world for a host of reasons, including possibly because of distortions such as those discussed here.4

The Brazilian government is taking steps to reformulate the role of BNDES. This is long overdue and much welcome. It is also a clear example of what the region needs: development institutions that fulfill their true calling, not a proliferation of institutions that occupy undue space.

Notes

1. ALIDE (Asociación Latinoamericana de Instituciones Financieras para el Desarollo) is Latin America's association of development institutions, comprising 121 members (first- and second-tier banks) across the region.

2. The detailed exercise will be presented in a forthcoming Policy Brief.

3. There is a growing literature that tries to measure the impact of distortions in capital markets on unobservable variables such as total factor productivity. See, for example, Z. Song and G. L. Wu, 2013, “Identifying Capital Market Distortions,” University of Chicago, Booth School of Business.

4. High reserve requirements are also the result of regulatory choices: Brazil's rigid regulatory framework reflects a history of disruptive banking crises that induced regulators to impose stringent capital and liquidity buffers.