You may remember the three-post series in 2011 summarizing the findings on the Impact of Economics Blogs here , here , and here . The final version, now titled “Quantifying Some of the Impacts of Economics Blogs,” is forthcoming in Economics Development and Cultural Change . The paper went through some changes as a result of the peer review process, all of which were for the better.
The main changes addressed comments that kept popping up repeatedly, be it from journal referees or from readers who would talk to us about these in passing. For example, we took out the non-experimental evidence on blogging improving bloggers’ reputations – there simply was too much room for unobserved heterogeneity to account for the finding. Also noted in the concluding section is another thing we still hear from readers of Development Impact (and was a favorite topic of some journal referees): blogs could have detrimental rather than beneficial effects for the blogger and his/her institution. Finally, because we don’t know your opportunity cost of reading this blog, we cannot interpret the changes in our readers’ attitudes and knowledge our findings as overall increases in knowledge or welfare.
But, the one new analysis we did to supplement our arguments was in response to the comment that blogs are no more than self-promotional tools. Just as advertising works for products, if the blogger keeps promoting his/her work, there will be some effect on how much of it is consumed – without any significant dissemination of knowledge or spillover effects. While the advertising analogy is fair, the description that development bloggers mostly covered their own work did not sound right to us. So, we did a little digging…
We have examined every post from 11 development blogs for the six-week period between September 1 and October 15, 2012. These included popular blogs from larger development institutions, such as Inter-American Development Bank’s Development that Works , Center for Global Development’s Views from the Center , World Bank’s Development Impact and Let’s Talk Development , as well as Oxford University’s CSAE Blog , Chris Blattman.com , Acemoğlu and Robinson’s Why Nations Fail , among others. We then categorized each post into one or more of the following four categories according to the contents of the entry: (Category 1) blog post discusses blogger’s own research paper; (C2) blog post discusses blogger’s own ideas on research, policy, impact evaluation, etc.; (C3) blog post discusses a research paper by someone else from the same institution; and (C4) blog post discusses a research paper by someone else from another institution (Note that because a blog post could be categorized into multiple categories, the percentages of posts in each category can add up to more than 100%.). Posts discussing non-development issues such as how great a blogger’s new desk is  were excluded. Our findings are reported in the table below.
We find that, of the 255 posts in these 11 blogs during this six-week window, only 36 entries discuss the blogger’s own research paper (14%), and another 29 mention research papers by someone else from the same institution (11%) – indicating that approximately three quarters of all blog entries contained no mention of research by the blogger or someone else from the blogger’s institution.* In contrast, 171 blog posts (67%) in this sample contain some discussion of a research paper by someone from another institution, including a non-negligible number of posts that contain a substantive discussion of a research paper rather than a simple link to the paper or a ‘cut and paste’ of an abstract. These results suggest that while there is a significant element of self-promotion and the promotion of affiliated individuals and institutions that is present in blogging, bloggers also disseminate information about new research, ideas, and issues in their fields of expertise.
Taking into account the caveats noted above, we finish the post with the concluding paragraph of our paper: “Nonetheless, our results show that there are a number of positive externalities from economics blogs that are unlikely to be captured by the blogger herself. The presence of these externalities, coupled with positive costs of blogging, suggests that there may be an undersupply of good economics blogs. Organizations and academic institutions, which are curious but unsure about the benefits of having more of their staff blog in their field of expertise, may provide some incentives for blogging, closely track the impacts on institution-specific outcomes of interest, and keep the subsidies for successful blogs and discontinue them for others.”
* These figures increase to 22% and 18%, respectively if we exclude Udadisi from the list, which accounts for a large number of posts in our sample (102 out of 255). Similarly, 78 out of 153 total posts (or 51%) contain a discussion of research papers by someone from another institution when Udadisi is excluded from our analysis.