By Anjum Altaf
I am happy to engage in a debate with the Center for Global Development on US aid to Pakistan. However, for me the issue is not aid to Pakistan or aid in general but the analytical validity of CGD’s recent reports. I argued that CGD’s 2011 report was advocacy, not analysis and based on a reading of a summary of the 2012 report I concluded it seemed no different.
CGD has responded to my criticism of the latter but has, in what I consider a handwaving style, ignored my central concern and resorted to diversionary arguments to mount a defense. Here, I aim to show why CGD’s case remains a weak one.
CGD’s first point is that their report has been criticized both in Pakistan and in Washington and “perhaps this is a sign we have done something right?” But could it not be equally likely that they have gotten something wrong? This tendency, to assume without further questioning the interpretations or outcomes that favor their position, exemplifies what I consider the principal weakness of their reports.
Their second point is to ask if I have actually read their new report. No, I have not. I made clear in my critique that I was responding to a summary published as an op-ed. A reading of an institution’s output cannot be demanded; it has to be earned. I did read CGD’s first report and found it very weak. The summary of the second did not suggest a difference sufficient enough to convince me that reading another report by the same authors would be a good use of time.
The third point is to allege that I have not read other related outputs by Nancy Birdsall, the lead author of the two CGD reports. Indeed, I have not but my aim was a critique of a specific op-ed, not of the entire oeuvre of its lead author. The best of authors have a few poor pieces to their names, just as the finest of actors sometimes deliver forgettable performances.
The fourth point is to say that since my “riposte includes too many points for us to address individually here, so perhaps its best for us to simply and clearly re-state the core messages of our recent report, which we did not have the luxury of summarizing in our recent op-ed.” As far as I am concerned, a repetition of the core messages does not take care of my critique that these core messages rest on a very weak analytical foundation.
The fifth point is to mention that the core messages incorporate some ideas that I had myself suggested in my critique of CGD’s 2011 report. This may or may not be true but in either case it misses the point. There are some commonsense ideas that should be part of any sensible strategy of assistance if its benefits can be shown to exceed its costs. But random inclusion of a few credible ideas does not by itself make an analysis credible. This much should be obvious to any academic.
The problem with the CGD outputs remains that they are non-analytical – they aim to build the best case for the predetermined position that US assistance to Pakistan must continue because Pakistan cannot be allowed to “fail”. Whatever concessions are required to sustain that case are made and whatever favorable assumptions are needed to bolster the proposition are adopted. The resulting arguments are often so tortured as to be beyond belief. The authors conclude that “the report speaks for itself.” To me, the summary suggests it is a weak report. If the model that supports it, and I am not convinced there is one, were to be submitted to a quality peer-reviewed journal, it is highly unlikely to be accepted. Let that be the test that resolves this disagreement.
Any credible analysis of aid to a country must demonstrate benefits with reference to the no-aid scenario and must take into account the incentives, both good and bad, that are generated by the nature of the assistance. I have made the point elsewhere that “Pakistan’s leaders can mismanage and pillage the economy knowing American policymakers have determined that Pakistan cannot be allowed to fail – which is great for the leaders but terrible for the citizens.” Nothing I have read in the CGD reports or their summaries convinces me that this does not remain the case.