Caught up on this recent NBER WP on labor productivity growth and industrialization in Africa by McMillan and @AlbertZeufack nber.org/papers/w29570
Offers a very useful overview of trends in manufacturing and structural transformation in SSA; worth quick 🧵 #EconTwitter
The paper uses a range of data sources, but the first is the Economic Transformation Database (ETD) including 18 SSA countries that allows for estimation of value added per worker across countries.
Estimates show that labor productivity growth has been 2.5% in SSA since 2000; this is mostly driven by shift of workers from ag to non-ag (i.e., structural transformation). Minimal contribution of within-sector productivity growth.
At the country level, this graph shows that is essentially no correlation between productivity growth from structural change (blue) and within-sector productivity growth (green). In contrast, in industrializing Asian countries, the correlation is positive (right graph)
Hence the puzzle: in Asia, the manufacturing sector becomes more productive, draws workers in from ag, amplifying the productivity effects; in SSA, this pattern is not observed
Why? Employment growth in SSA manufacturing dominated by small / informal firms (<100 workers) w/w low productivity. Larger firms appear to be relatively productive, but they are not growing. This graph shows ⬆️ in manufacturing emp in 8 countries w/formal emp relatively flat
Again, in contrast to the pattern in Asia, where total + formal manu employment increase in tandem.
Paper also has a detailed case study comparing Ethiopia vs. Vietnam manu, highlighting key differences: much ⬇️ share of informal emp in Vietnam, much ⬆️ presence of foreign-owned firms, more “huge” firms (5,000+ emp): in 2017, >100 foreign firms of this size in Viet, 0 in Eth
Overall, very interesting and thought-provoking set of stylized facts to put structural transformation in SSA in context of patterns observed elsewhere (will also pitch recent paper w/@BilgeErten unpacking the role of export-driven growth in stimulating structural transformation
Flyouts are starting, so here’s a quick 🧵 on advice for introverts like me. One of the challenging parts of the jm is high social interaction, possibly made more difficult if you have constraints (familial, locational) that you want to keep private at first. #EconTwitter
As to strategy for disclosures – I’ll let others speak to that, other than to say I agree you should always be truthful, but you can choose not to reveal certain things. But that can add stress, making it even harder to chat comfortably.
So, a few quick thoughts (more on the side of the non-econ part of the discussions, not the research part.) Come prepared with a few topics that are of broad interest. Hobbies? Books? Movies? Food tastes? All good. Don’t force it, of course.
As new PhD students start to look forward to their first year, short 🧵 on challenges in collaboration in grad school (and its potentially gendered dimensions).
Many people advise grad students to rely on their classmates: first in coursework, later on projects / as coauthors.
I endorse that advice! But it can also be hard to follow. I attended two grad programs (MPhil and PhD) and had similar experiences in both. There were large, energetic, overlapping-networks problem set groups that formed quickly.
They were mostly dominated by men (unsurprisingly; econ grad programs are mostly dominated by men) and, to describe it neutrally, had a fast-paced style. Always an introvert who was becoming more so, I was uncomfortable and anxious about trying to participate.
Enjoyed the presentation by @elianalaferrara today at World Bank DIME of work joint with Baumgartner, Rosa-Dias, Breza and my awesome coauthor Victor Orozco: evidence around a peer education program targeting early sexual activity teen pregnancy in Brazil.
The authors have a fascinating evaluation comparing a peer educator program with three alternate selection mechanisms for educators (school-driven; selection via peer nomination of popularity; selection by centrality in a formally mapped network) to a control arm.
In general, the peer education program is very effective: ⬆️ knowledge and communication around sexuality, contraceptive use; ⬇️ teen pregnancy. The peer educators chosen by schools (the default method), however, were generally ineffective!
So much of what we hear around RCTs are exciting stories about how evidence is used to inform policy. Which is awesome! I love evidence-informed policy. However, I'm sure many of us have also had experiences that are different, and more challenging.
In the spirit of transparency, wanted to share some different (anonymized) stories about use of evidence. Short 🧵
#1: program has mixed effects (largely null for downstream outcomes). Funding for the implementer concludes, implementer and funder move on. What happens? Brief discussion, draft paper shared (0 replies), almost no policy learning (hopefully research community benefits).
Lately I’ve been thinking more and participating in various conversations about how USAID commissions, uses research. Huge topic! But wanted to do a short🧵on what I’ve learned. 1/n
First and foremost in the hearts of most economists is DIV. DIV is awesome, as many others have pointed out! See this recent blog by @DaveEvansPhD and colleagues 2/n cgdev.org/blog/case-evid…
But, DIV primarily funds evaluation of pilots and other interventions that are implemented outside of USAID and are not directly related to the work of missions – as summarized above (there are some exceptions). In that sense it is often separate from the main aid portfolio 3/n
After doing a lot of reformatting to meet a journal page limit (which, TBC, I support), I started to wonder - why don't journals impose limits on the referee reports that lead to these long papers? E.g., 1-2 pages; or alternatively, 3-5 (choose N) substantive suggested revisions
Seems like this could help with a lot of problems - long review times, tedious revisions, bloated papers that are hard to read, indigestible appendices, etc. Hard to enforce, but editors could suggest that material beyond the limit would be ignored.
Plus, I suspect many referees would be very happy for explicit guidance that allows for coordination, since (many of us) are concerned about the perception of submitting an un-thorough or low-quality report relative to others. . .