A couple of weeks ago, Brazil’s Federal Police arrested 46 people in 10 states in an operation that accused a construction company and public employees of corrupting procurement transactions for infrastructure projects. Under suspicion of being involved, the Minister of Mines and Energy resigned last week and just yesterday the supreme court ordered the deputy head of the Federal Police to step down, also accused of having a role. Last year, a corruption case that involved overpriced municipal purchases of ambulances from another private group resulted in four congressmen losing their mandates.
As with the case of informality (theme of my previous post), an obscure or corrupt government procurement system has the potential to harm fair and transparent domestic market competition and, in doing so, reduce the incentives for productivity growth. One estimate of the size of government procurement in Brazil puts it at 7% of GDP in 2002, roughly half the export market in that year. This number may be considerably underestimated, however, since it appears to not include all state enterprise purchases as well as all municipalities. Internationally, government procurement seems to typically account for over 10% of GDP (http://www.cid.harvard.edu/cidtrade/issues/govpro.html). In any case, if foreign markets demand competitive firms, government procurement could play a similar role for another chunk of the private sector more focused on the domestic market…or not.
Brazil’s government procurement law is from 1993 and, according to a 2004 World Bank report, its emphasis on formal requirements is responsible for pushing most procurement processes away from competition, “in order to avoid the time consuming bidding process.” In 2000, an internet accessible reverse auction system was launched and in 2005 the use of this system was made mandatory for off-the-shelf goods (www.comprasnet.gov.br). It is considered to be highly successful in that it reduced average costs of good purchased as well as the time required. Infrastructure and other types of projects, however, are not covered by this system. This may change soon. In January, as part of its announced Growth Acceleration Plan (mostly infrastructure investments and theme for a future post) the government included modifications to the 1993 procurement law and that are currently being discussed in the senate. Among these modifications, procurement for infrastructure would be included in the electronic system up to a certain BRL value. However, exactly what modifications will come out of the current political process is still not clear, since the government’s proposal was significantly modified by the house before following to the senate. Another modification to Brazil’s government procurement system will soon be brought by the General SME Law approved in December of last year. The law allows for SME set asides in procurements up to BRL 80,000 and allows the government to require of its suppliers to include SMEs as subcontractors.
Whether the modifications being done to Brazil’s government procurement system will promote greater transparency and broaden competition remains to be seen. I have chosen to highlight the current process under the assumption that it matters. Explicitly, the assumption I’ve been working with is that fair competition in demand markets and broad access to factor markets are key to productivity growth while also likely being “pro-poor.” The same assumption guided my choice of informality as theme for last week’s post, a post that led to interesting follow-ups by Nouriel Roubini and Otaviano Canuto, and to a brief discussion on the relevance and limitations of cross-country comparisons in helping us identify what constraints to growth are binding and what are not in any particular case.
On the subject, an interesting recent paper was brought to our attention by a commentator to Otaviano Canuto’s post, Bruno Saraiva. The paper’s authors, Wendy Carlin and Paul Seabright discuss limitations to inferring from cross-country regressions, subjective surveys or case histories, what business environment policies would have a greater impact on productivity growth, and they suggest it is possible to combine one or more approach to improve recommendations. Even with this approach, however, there seems to be ample room for different interpretations of the evidence and, as the authors themselves recognize, this still does not tell use what should be done, since costs must still enter the equation and some constraints may be beyond the policymakers reach to affect. I would add that this analysis also doesn’t tell us what can be done, since the various and conflicting interests and implementation issues that conform the realm of the policymaker must also be included in the picture. Otaviano Canuto touched on yet another element to be factored in regarding policy – the dynamics of reform – when he noted the importance of sequencing. I believe all these are important issues and look forward to a continued productive discussion.