Holding the World Bank Accountable for its Research:
The Case of NAFTA
May 24, 2007
The recent controversy that led to the resignation of World Bank President Paul D. Wolfowitz highlighted the lack of accountability at the World Bank. But it is not only its governance that suffers from the fact that the Bank, with 185 member nations, is ruled primarily by the U.S. Treasury Department. There are also serious problems with the Bank's research.
Last year, the World Bank established a panel of economists to evaluate its research from 1998-2005, including nearly 4,000 papers, books, and reports. Among other problems, the panel had "substantial criticisms of the way that this research was used to proselytize on behalf of Bank policy, often without taking a balanced view of the evidence, and without expressing appropriate skepticism."
But sometimes the problems are even worse, as when the Bank publishes erroneous research results, which influence important policy debates, and then refuses to correct its errors. The Center for Economic and Policy Research (CEPR) has tried to hold the Bank accountable for its research. In December 2003, the World Bank released a paper which purported to show that NAFTA had a positive influence on Mexico's economic growth. The paper was timed — and indeed, rushed — to have an influence on the political debate. It was released on the final day of the CAFTA negotiations and coordinated to coincide with the 10th anniversary of NAFTA (January 2004), which brought a spate of newspaper articles on the agreement's impact.
The World Bank's results, now shown to be erroneous, influenced the public debate. For example, they were cited by a Washington Post column in December 2003 and the Washington Post editorial board in January 2004, in support of the newspaper's arguments for CAFTA. In March 2004, CEPR responded with a paper that highlighted errors in the paper that negated its results.
The correspondence and papers below, between CEPR and the World Bank, explain the details of the Bank's errors and why they matter. The bottom line is that, because of uncorrected errors in the Bank's analysis, their econometric results cannot support the claim that NAFTA had a positive effect on Mexico's growth rate. As of today, May 24, 2007, the Bank has still not retracted its conclusion or corrected this mistake.
December 2003 — The World Bank publishes NAFTA and Convergence in North America: High Expectations, Big Events, Little Time (as Chapter 1 of Lessons from NAFTA for Latin America & the Caribbean).
March 2004 — CEPR publishes NAFTA at Ten: The Recount, which explains problems with the World Bank's methodology in NAFTA and Convergence in North America.
April 2004 — The World Bank posts a second version of NAFTA and Convergence in North America.
May 2004 — The World Bank posts a third version of NAFTA and Convergence in North America.
October 2004 — CEPR publishes Getting Mexico to Grow with NAFTA: The World Bank's Analysis, which explains the World Bank's methodological problems and errors that were not addressed in its subsequent versions of NAFTA and Convergence in North America.
November 2006 — The World Bank publishes The World Bank's Research on Trade Policy, 1998-2005: An Evaluation in which UCLA economist Sebastian Edwards reviews Lessons from NAFTA for Latin America & the Caribbean.
January 2007 — CEPR sends letters to Sebastian Edwards and François Bourguignon (World Bank's Chief Economist) reiterating the original concerns with NAFTA and Convergence in North America and asking for a correction.
March 2007 — CEPR receives a reply from the authors of NAFTA and Convergence in North America, and from Bourguignon, who defers to the authors. The reply defends thepaper and does not acknowledge any errors.
April 2007 — CEPR responds, explaining again the errors that negate the World Bank paper's conclusions, and once again asking for correction.
For more information, please contact Dan Beeton at 202-293-5380 x104.