Links round-up

Hi all,

 Skipping all the usual jokes, I’m just going straight into a rant here: when we turn our politics into a series of vitriolic attacks on those who don’t share our views, demonise and dehumanise those who don’t share our language or (proximate) origins, and treat dissent and political discourse as a battleground rather than an exchange of ideas and evidence (not stats, not numbers, not claims, but evidence), then we’ll always run the risk of winding up where we did yesterday. I’ve been guilty of this before, and only a couple of days ago got a (fully deserved) dressing down from a colleague for talking down about people who don’t share a particular political view I hold – I hope I take that to heart. There have been various lovely tributes to Jo Cox, including a number on my Facebook feed, as she was known and loved by many people who worked in DFID and development. I particularly like this one from Andrew Mitchell. It’s all very distressing. And disappointing.

 1.       Staying on UK affairs for a moment, here’s Ben Casselman at FiveThirtyEight on what Brexit could mean for the economy. It’s balanced and accounts for the views of both sides of the debate. The spread betting on the outcome of the referendum remains razor thin so if you have views on this, get out and vote next week, come rain or shine. If you think it won’t affect you – read more about it. It will, so be as well informed as you can be.

2.       Dan Honig at John Hopkins sent an e-mail to Matt Collin (CGD) and I last week, about the relative merits of evaluating development projects using ex ante, pre-determined criteria vs. using an observational approach that decides what to consider after the fact, setting off a geeky, ranty back-forth-and-back-again between the three of us, which Matt then published. While the discussion of economic evaluation is riveting (of course it is, given how handsome and brilliant the three protagonists are), the most notable point made in the discussion is when I turn my prognosticative skills to the basketball, and declare “[the] Cavs might still push it to 6, and if they do, Lebron is perfectly capable of turning in two monster performances in a row”. Here’s the first of the two – Sunday night, we’ll find out if the second materialises…

3.       David Evans was at DFID earlier this week giving a brilliant presentation on one of his recent projects, looking at management capacity and quality of healthcare in rural Nigeria. Earlier in the week he attended the Annual Bank Conference on Africa in Oxford, which had an urbanisation theme, and together with Markus Goldstein (who also organised a great presentation here a few months ago) wrote a summary of every paper presented there. Strongly encouraged for all economists, particularly those working on economic transformation. Some of this research is fantastic, and providing new insights into how cities can impel or impede development.

4.       I’m in Paris at the moment (mangling the language in my interactions with the locals) and have been at the OECD today, presenting on the need for a more thoughtful and nuanced approach to thinking about development and migration. I don’t think I would have gotten away with Chris Blattman’s latest blog

5.       I really liked this: Owen Barder and Matt Juden dig in to the old chestnut that middle-income countries get more aid than low income countries, and find (like the t-shirt I keep wishing existed says): ‘it’s a little bit more complicated than that’. In fact, by most intuitively reasonable measures, the opposite is true.

6.       I am writing this e-mail with my feet planted on the table, leaning back in my chair, having been told that this is a ‘power pose’, albeit one that makes it very difficult to type. Tim Harford gives the idea a shoeing, while making a broader point about the reproducibility crisis in the social sciences, ending with a rule of thumb by Andrew Gelman, which is always a good idea. Somewhat related: David McKenzie points out that, numerically at least, RCTs really haven’t taken over development research.

7.       And finally, because I started the e-mail on a sombre note, let’s end it more upbeat: here is a statistical analysis of every movie The Rock has made, noting his inexorable rise and increasing awesomeness. “[He] manages to say the line ‘F**king centaurs,’ and you buy it. Do you realize how hard it is to sell the line ‘F**king centaurs’?”

 Passez un bon weekend, tout le monde!


Posted in Uncategorized | Tagged | Leave a comment

Links round-up

Hi all,

 Even though it’s been almost a week and Sri Lanka are playing in the cricket right now (and – dare I say it – looking pretty darn competent), this week’s preamble is all about The Greatest. Last Saturday, after hearing the news, I re-watched When We Were Kings and read all the tributes I could. LitHub did a nice summary of the best books about him and  Vox did a rundown of the best publically available writing. For the fans of the actual boxing, here’s his fight against Cleveland Williams, probably his finest performance, at the peak of his physical abilities. My own favourite story about him: When I worked in Zanzibar, a portly, middle-aged man came into my office looking for my (absent) office-mate. He introduced himself as the Manager of the Zanzibar Ports Authority, and asked me to say he had come in to say hello. I asked him what his name was, and he flashed me an enormous smile and started swiftly dancing around the office, shadowboxing with dizzying speed, before withdrawing a business card and announcing himself as Mohamed Ali. I’ve always wondered if he did that every time someone asked him his name. I’d be tempted to.

 1.       “Today on the show, we meet the man who stole my door and who gave us the hell that is the open office.” One of the great joys of working in the Chief Economist’s Office is that we occupy an odd little corner of the building which is difficult to find and relatively small. As a result, though it’s technically an open plan, it feels more like a small shared office space. Planet Money look at the invention of the open plan, and finds the architect responsible for it (transcript). Almost everyone hates the open plan (Shalom Auslander describes it as like ‘being inside a migraine’), though in some contexts it is beneficial to team-level output, but they nail the appeal: it looks cool, and it’s cheaper.

2.        I keep banging on about the importance of raising the returns to work in developing countries. The problem is not that people don’t have enough work, nor that they aren’t enterprising enough or anything like that – it’s that all the structures that allow one to do well and maximise welfare by selling or using your labour are sub-optimal. A nice new paper reinforces this deeply held prior: people in developing countries actually work more than their counterparts in rich countries, and do so for far, far lower returns. The global incomes gap is thus even larger when we account for hours worked.

3.       Speaking of the global incomes gap, here’s Branko Milanovic reviewing a book with a new(ish) explanation of the Great Divergence of the west (and specifically Britain) from the rest of the world, focusing on the role of the interventionist state. The book also goes on an eminent academic killing spree, attacking Jared Diamond, Greg Clark, Acemoglu and Robinson, Kenneth Pomeranz and David Landes. He also made me guffaw by describing Andre Gunder Frank’s ‘usual lack of nuance and overdose of self-assurance’.

4.       Two bits of good thinking on migration from CGD. First Hannah Post and Owen Barder carefully look at the numbers of ‘refugees’ and ‘asylum seekers’ in Europe and find that rhetoric and reality do not necessarily match; and then Lee goes on a longer version of his earlier rant against the ‘worst use of aid money ever’: “Whilst the EU apparently trusts the government of Sudan to respect the human rights of foreign refugees, the same Sudanese government is simultaneously bombarding its own citizens…”

5.       “The new approach to economics should include two different kinds of theories: normative models that characterize the optimal solution to specific problems and descriptive models that capture how humans actually behave.” Richard Thaler on the state of modern economics and the direction it needs to take, arguing that everything has a behavioural element, and that the sooner we internalise that into mainstream economics, the sooner there ceases to be a separate discipline of ‘behavioural economics’.

6.        I don’t want to be a total downer about John Oliver buying loads of debt and then forgiving it, but as Tyler Cowen might say: solve for the equilibrium. That’s not to say I think the current situation is fine, either – with vulture funds buying loads of debt cheaply to try and squeeze blood from people, I don’t think the direction those incentives push us is any better. Rather, I think that grand gestures are just that: nothing more than a gesture. The problem is financial education, understanding of risk, and the need for a really open discussion about how far we are willing to trade off risk and return among the poor, and how to mitigate or insure risks.

7.       And Lastly, after the musical end to last week’s links, someone sent me this gem: A North Korean pop group singing their great hit, Let’s Support our Supreme Commander with Arms.

 Have a great weekend everyone!


Posted in Uncategorized | Leave a comment

Links round-up

Hi all,

 So, in cricket news, shortly after talking about how Sri Lanka’s performance in the second test was making me cheerful, they went and collapsed to 101 all out. I await the Lord’s test with such fear and trembling I may need to call in sick (note to my line manager: if I actually do call in sick on those days, it’s a complete coincidence and I actually am sick. Either that or Sri Lanka are actually in a position to win the test and I’m attempting to scale the outside wall in St. John’s Wood to get a glimpse of the winning runs, in which case it’s entirely excusable). Right, with that potentially career-limiting preamble out the way, on to the economics.

 1.       This isn’t really economics (ok, it isn’t economics at all), but it does involve some very nifty statistics. With LeBron James very unlikely to win his third championship this year it’s worth remembering: victory is quite a blunt proxy for brilliance. In dragging what have sometimes been frankly mediocre teams into the finals of the NBA Playoffs, and sometimes putting up historically brilliant performances in losing causes, he may have already made the case for being one of the greatest players ever. Of course, we’re not going to appreciate this until at least ten years too late.

2.       Two perspectives on the evolution of the IMF. Ostry and co-authors, all economists within the Fund, do something very brave here, arguing that the neoliberal dream sold by the Fund in the 80s and 90s was oversold and that even in its great success, Chile – might not have been the fundamental reason for success. This is all part of a general trend in the Fund of rowing back on neoliberalism and conditionality in programme lending, at least in rhetoric. However, three political scientists and sociologists argue that that’s all this was:  a rhetorical flourish, showing a marked increase in conditionality after 2008, bringing it right back up to its historical high-water point. I find this criticism underwhelming. Firstly, I can’t be the only person who can think of an event in 2008 that might have something to do with the need for a lot of mandated structural reform (hint: it rhymes with ‘Brobal Binancial Brisis’). Secondly, their argument that countries need the flexibility of policy making in order to recover from crises misses the point – in many places, political capture and rent-seeking have led to the establishment of terrible economic institutions that might require an external actor to remove. There are plenty of reasons and ways to criticise the blanket application of conditionality and neoliberal orthodoxy, but the worst is to do so with the blanket (unspoken, but present) assumption that they’re always wrong.

3.       I love it when Branko Milanovic goes off on philosophical tangents about the meaning of economic or pseudo-economic terms, like ‘perfectly unproductive labour’, as he does here. In doing so, he takes us on  a fascinating trip around the history of economic thought, here touching on Marx and Adam Smith. I love that Branko both understands Marx intimately and isn’t afraid to discuss him completely seriously, risky business for economists these days.

4.       Chris Blattman reproduces this brilliant graphic from Stefano DellaVigna and Devin Pope, summarising the results of their investigation into the effects on effort of encouragement and financial incentives (write-up). They find that substantial financial remuneration, linked to performance stimulates much higher performance. But before you despair that we’re all greedy rationalists, here’s a wrinkle: the task they set the people they experimented on was perfectly menial (unproductive?) work – cross-country regression analysis (just kidding, it was typing ‘a’ and ‘b’ repeatedly in alternation). I suspect that the results would be very different for work which has strong intrinsic motivation, like economics. Or cricket.

5.       The rise of renewables have made the existing grid infrastructure for energy provision in most of the developed world somewhere between ‘slightly’ and ‘hopelessly’ out of date. The Western part of the US is closer to the ‘hopelessly’ side of this spectrum. Vox investigates. One of the very exciting possibilities in development is that someone will properly fund an absolutely cutting edge energy provision grid in a developing country, one of the rare occasions where a poor country can immediately leap to the technology frontier.

6.       The basic income referendum in Switzerland is scheduled to a resounding ‘no’.

7.       And finally – I would consider any of these popular wedding playlist songs grounds for an instant annulment, except number 7, though I wonder if all the people playing Billie Jean at their wedding have ever listened to the lyrics. “The kid is not my son” has to be close to the top of the list of things you should never say at a wedding. Via Tom Coward, he’s a better song than most of these, with an amazing video. From the DRC: Karibu ya Bintou (again – not a wedding song…)

 Have a great weekend, everyone!


Posted in Uncategorized | Leave a comment

Links round-up

Hi all,

 Though I’m technically off work today, celebrating the Queen’s birthday (I’ve spent it wearing a bejewelled hat and sitting in a gold chair, thinking about poverty), I decided to log on to write this week’s links: Sri Lanka’s not-historically-appalling start to the second test against England (anyone who mentions the first test gets cut from this mailing list – so be warned, those of you crafting clever insults) has put me in a good mood, and I’ve spent the last couple of hours in the park, lying in the sun and reading. What a day.

 1.       Since I’m in a good mood, let’s start with something happy. How often does a DFID-funded (or part-funded, I’m not hugely sure) get a glowing representation in the popular media? Well here’s Planet Money doing their thing with the YouWin! Competition in Nigeria (transcript), which has featured in these links before. To recap: David McKenzie ran an RCT giving large grants of around $50,000 to Nigerian entrepreneurs – selected largely at random – and sat back to monitor the results. They were good. How good? Chris Blattman suggested it might be the best development intervention ever (I still love David’s chuckling response to that: “ha, nooooo. That’s migration.”) The NPR guys interview David, Chris, Ngozi Okonjo-Iweala and, most importantly, one of the winners, Lariat Alhassan, and paint a glowing picture of the work (as an aside, read David’s full paper – it’s amazing).

2.       Speaking of YouWin! David has milked another brilliant paper from the data: he surveys both the competition winners and those who weren’t successful, and asks them how much additional success they think the winners got as a result of winning the competition. Interestingly, both the winners and the rest of the field dramatically overstate the gain to winning the competition. This is a nice demonstration of a broader point: firms aren’t always the best judge of what their biggest constraints are, or how constraining they are.

3.       About two weeks ago, CGD asked what the best way  to allocate refugees around the world – looking at this as an issue of equity. Tim Harford reports on a really interesting alternative approach: looking at it as a question of maximising efficiency, and reports on a really interesting idea, using matching models to allocate refugees to places. These models look at what the refugees skills and contribution might be (on an individual/family basis) and what their needs are. It then looks at what the needs and capacities of different potential locations for them are, and devises a way of matching refugees to location that maximises collective welfare. The lovely thing about this approach, as he notes, is that it is completely independent of the politically toxic question of how many to accept – and so the piece is a nice complement to Nancy et. al.

4.       So this, definitely isn’t going to end badly: an Israeli firm has come up with facial recognition software to ‘predict’ murderers, paedophiles and terrorists. I’m going to call bullshit anyway, but even if they have, I’m fairly sure a whole bunch of dystopian novels start from this premise.

5.       A really good profile of Dani Rodrik. I don’t always agree with his papers, but he consistently asks some of the most interesting questions in economics, and the profile gets it right: he’s a born dissenter, never just willing to go with popular opinion without giving it a close reading first.

6.       This week in over-sharing meets data journalism: an American blogger tracks every the time she cried over a year, noting the duration, intensity and cause of the weeping. It’s interesting in a creepy, voyeuristic kind of way.

7.       Lastly – an absolutely amazing, moving account of a Japanese translator’s mission to create a Japanese-language version of Stoner by John Williams, after being diagnosed with terminal cancer. He died after completing all but the last page. Read it.

 Have a great long weekend, everyone!


Posted in Uncategorized | Leave a comment

Links round-up

Hi all,

 Well it’s been a completely  manic week, so if these links feel a bit anaemic and tired, rest assured that they reflect their maker. And if they seem on the brink of absolute despair and weeping, blame Sri Lanka, whose approach to scoring might be described as binary, except that would indicate a few ones among the ducks. Now please excuse me while I go and punch a wall.

 1.       This week, Dietrich Vollrath used Spinal Tap to win an argument about economics (and trust me, this is a win. It’s the equivalent of the Cavs versus the Raptors in Game 1 – a total and complete smackdown). The argument was about whether getting better at doing business, as measured by the Doing Business Indicators can improve growth. Dietz breaks this idea apart carefully. He looks at the way in which the Doing Business scores are put together (old hat to DFID economists); then he looks at the rate of growth you can achieve even if doing business is your constraint to growth, and finally, he suggests that one of the reasons for Singapore’s success isn’t so much it’s business environment as the fact that it’s a single city – and measured against other cities it doesn’t actually look that impressive. The summary does little justice to all the wisdom in the post. Dive in.

2.       (So, Sri Lanka just lost another wicket. We should just declare now and stop embarrassing ourselves). Anyway, while I tear my hair out, read this superb discussion of inequality from Branko Milanovic. He looks at firm structure and how the decline of huge companies that employ loads of people has actually led to an increase in inequality. This seems counterintuitive – surely smaller companies should be less unequal? But no – Branko doesn’t make this connection, but one of the best papers of recent years, Nick Bloom (and co-authors) piece on firms and inequality demonstrates this: most inequality is now between firms, not within them. Basically firms are specialising more and more, and the lucky people who work for the really productive ones earn heaps – but within the firm, there’s not much inequality. For the rest of the people, though, incomes are falling further and further behind, as they are no longer hooked to their more productive peers (as a concrete example, think about the rise of outsourced security guards, who are no longer employed by the same firm as the buildings they guard).

3.       Who would have thunk it? It turns out we actually kind of need low-skilled migrants, and without them a lot of firms will find it very hard to fill jobs.

4.       Every time Markus talks about research methods we should ask sit down, shut up and listen. Here he gives really interesting tips on how to run research in conflict-affected areas. This is interesting even for those of us who don’t commission or run research projects, as an insight into the amount of thinking and care that goes into the design of a good piece of field work.

5.       Last week, I linked to Harry Enten’s piece on why it was so hard to predict Trump. This week, it’s Nate Silver’s turn, and he’s typically thorough, and proposes a good way of doing better next time, again turning to Bayes.

6.       Lastly, apropos of absolutely nothing, here are two great pieces on Sumo wrestling – which for some reason spawns really high quality writing. First, FiveThirtyEight uses a rich data source stretching back aroud 200 years to ask if Hakuho, the current top dog, is the greatest ever (the article is fascinating, even if you nothing about the sport). And older, but brilliant – Fightland on Vice break down what the heck Sumo is, for people like me who thought it was just plus-sized dudes bouncing bellies.

 Have a great weekend, everyone!


Posted in Uncategorized | Leave a comment

To What Extent Resource Discovery Influences the Politics of Fiscal Decentralization

“North Sea oil and gas reserves are another matter of much dispute. Mr Salmond says an independent Scotland would earmark a tenth of revenues – which the Yes campaign puts at about £1bn a year – to form a Norwegian-style sovereign wealth fund, creating a £30bn pot over a generation. Prime Minister David Cameron says the North Sea has been a British success story – and now oil and gas are getting harder to recover it’s important to back the industry with the “broad shoulders” of the UK.” Barford (2015)

Haggling over a fair share of the tax revenue between the central and the provincial governments is often an integral part of the political theatre in many countries. Natural resource revenue for its part plays a crucial role in influencing the politics of revenue sharing. Whether the issue at hand is fiscal federalism or autonomy or secession, the geographic location and the distribution of natural resource revenue seems to play a role. For example, the discovery of North Sea oil off the coast of Scotland has underpinned the Scottish case for sovereignty since the 1970s. North Sea oil was an integral part of the political discourse on either side of the Scottish independence referendum debate in the UK in 2015. The same could also be said about the secession movement of the three mini Indian states of Jharkhand, Chhattisgarh and Uttarakhand. These three mini states split from the three large states of Bihar, Madhya Pradesh, and Uttar Pradesh respectively in the year 2000 and they also happen to be endowed with one of the largest mineral deposits in the country. Bolivian indigenous communities of the Aymaras and the Quechuas not receiving a fair share of the natural gas revenue sparked mass protests and political instability in the country which led to the nationalisation of gas fields in 2005.

The UK's 2015 Oil Reality (Creative Commons)

The UK’s 2015 Oil Reality, by Michael Elleray (Creative Commons)

In spite of the potential connections, research on the interrelationship between natural resources and fiscal decentralization remain rare. Standard models of fiscal decentralization assume benevolent governments at the central and regional levels. They maximize the sum of utilities of residents in their jurisdiction and provide local public goods. Therefore, there is merit in fiscal decentralization or centralization depending on the nature of externality that the provision of local public goods generates for other regions in the country. Alternatively, another class of models view fiscal decentralization from a ‘Neo-Hobbesian’ perspective whereby the central government is a revenue maximizing Leviathan only constrained by the constitution and bottom-up democratic pressure via the regional governments. Under both of these approaches, one would expect the spatial distribution of natural resources and the quality of political institutions to matter by influencing the power relationship between the central and the regional governments. Yet studying the effects of natural resources and political institutions on fiscal decentralization remains on the periphery of this literature.

The recent CSAE working paper of Sambit Bhattacharyya, Louis Conradie and Rabah Arezki systematically explores the causal effect of natural resources on fiscal decentralization and how the quality of political institutions affects this relationship. In particular, the study exploits the exogenous variation in giant and supergiant discoveries in oil, gas and mineral reserves to set up a quasi-natural experiment to identify the effect of natural resources on fiscal decentralization. The study analyses the effect of resource discovery as an exogenous news shock using a global dataset covering up to 77 countries over the period 1970 to 2012. Furthermore, the study also estimates the effect of resource rent on fiscal decentralization.

The study makes the following contributions. First, establishing causality is the main motivation in this literature and the study presents a credible strategy to achieve that objective by using the exogenous news shock of resource discovery as an identifier. Second, the study uses a novel geocoded dataset on resource discovery. In particular, the new dataset is able to distinguish between 11 different minerals (gold, silver, platinum group elements (PGE), copper, nickel, zinc, lead, cobalt, molybdenum, tungsten, uranium oxide) and oil discoveries. Third, the study is first to analyse the effect of resource discovery on the politics of fiscal decentralization. In particular, it explores how democratization influences the relationship between resource discovery and fiscal decentralization.

There is no obvious prior when it comes to the effect of natural resources on fiscal decentralization. On the one hand resource discovery could embolden a central government who is acting as a revenue maximizing Leviathan to act far more unilaterally and centralize fiscal affairs. On the other hand resource discovery could also incentivize the central government to decentralize in order to either expand political patronage or improve the efficiency of public spending by addressing the preference matching problem. (Note that the preference matching problem refers to the mismatch in preference between the local population and the rest of the country with regards to public spending and revenue collection.) Therefore, the lack of a strong prior either way makes this a valid empirical question.

Estimating a model controlling for country specific unobserved heterogeneity and trends, time varying common shocks, discovery history in the previous decade, GDP per capita, and heterogeneity in the measurement of fiscal decentralization the study finds that resource discovery (both oil and minerals) has very little effect on fiscal decentralization from the revenue side. However, the former appears to induce centralization on the expenditure side and the effect seems to be driven by oil discovery and not minerals. The intertemporal effect of resource discovery on revenue decentralization (measured by revenue share) also appears to be statistically insignificant both 10 years pre-and-post discovery. The estimated coefficients however indicate expenditure concentration up to 6 years post discovery. The study also document that permanent democratization and the quality of political institutions have a differential impact on the effect of resource discovery on fiscal decentralization. In particular, the study finds that resource discovery leads to fiscal centralization in locations which have not experienced permanent democratic transition.  This effect is primarily driven by oil discovery. The study notes similar but statistically insignificant trends with mineral discovery.

Which fiscal institutions respond most to the resource discovery news shock and democratization? The study finds that tax and intergovernmental transfers respond most to the shocks of resource discovery and democratization. The institutions of own source revenue, property tax, educational expenditure, and health expenditure do not seem to be significantly affected. The discovery news shock might affect government revenue and spending through expectations but any direct effects on revenue collection have to wait till the start of production. The marginal impact of resource rent could be much more immediate and direct. Using both the standard fixed effects model and the instrumental variable (IV) method the study finds higher resource rent leads to more fiscal centralization and the effect is moderated by permanent democratization. This pattern is observed for both oil and mineral rents even though the effect is albeit weak for the latter.


Dr. Sambit Bhattacharyya is a Senior Lecturer in Economics at the Department of Economics, University of Sussex. His research interests are in the areas of development economics, economic history, and political economy. He has authored a book entitled Growth Miracles and Growth Debacles: Exploring Root Causes and published widely on topics related to natural resources and economic development.


Barford, V. (2015). Scottish Independence: Five Unresolved Questions, BBC Online, 25 August, 2015.

Bhattacharyya, S., L. Conradie and R. Arezki (2015), “Resource Discovery and the Politics of Fiscal Decentralization,” CSAE Working Paper Number 2016-5, University of Oxford.

Posted in Uncategorized | Tagged , , , | Leave a comment

Links round-up

Hi all,

It’s Friday the 13th, so obviously the Spurs lost. How else can it be explained? I’m still in shock. After the 27-degree sunshine in London (for which, of course, I was in rainy and chilly Addis Ababa), Leicester City winning the premiership, and the continued rise of Trump, this is clearly a sign of the end of days. So quickly, to the economics, before the coming apocalypse makes the Solow Growth model and convergence of incomes moot.

1.       Speaking of Trump, I think I need to eat some serious humble pie. After the first primary, I remember confidently stating that there was no way on earth that he could possibly win the Republican nomination; that early polls are usually wildly inaccurate; that no-one with such net negative favourability ratings had ever won. Well, he did; and we need to work out why so many people got this so wrong. The best piece of navel-gazing comes from one of the more consistent Trump sceptics, Harry Enten at 538. I am now resolutely making no predictions whatsoever, because my ‘put your house on the Spurs’ statement indicates that my endorsement is a kiss of death, and I’m not taking the blame for President Trump.

2.       In another sign that the Idiocracy is almost upon us, an Italian economist was chucked off a plane for writing in ‘terrorist code’. The said code? Algebra. Granted, the Lagrangian Multiplier is probably one of the scariest things I’ve seen, but still, make no mistake, this is the worst. I mean, what’s the reasoning here? Person writes in non-English script, must be Bin Laden? Even if Guido Menzio was Middle Eastern and writing in Arabic, how did it get this far? Suspicion was raised because he was not white enough and was writing in a foreign script. This is outrageous.

3.       Perhaps the only silver linings here are that Guido’s work is getting more attention, and the Economist did something funny (yes, I know – end of days, see the intro) with their list of ten signs that you’re sitting next to an economist on a plane. If you’re sitting next to this economist, you can tell by the loud snores and constant, annoying, fidgeting.

4.       The SDAs on the list might not like this, but here’s Stephen Porter arguing that qualitative research is increasingly irrelevant. Before the rotten tomatoes are launched my way, I think qualitiative research is really important (and always make this point in trainings on using evidence effectively), but he’s right that there’s a trend among some researchers into wilfully esoteric and vague work, which makes it less attractive to policy makers.

5.       And, for fairness’s sake – here’s Dave Evans on a bugbear of mine about quantitative research – how little of it provides information on cost and cost-effectiveness of interventions. This always annoys me – it matters if something is really expensive! Speaking of research – here’s some pretty terrible practice – a researcher released personally identifiable data on 70,000 OKCupid users, and then casually brushed it off. Gah.

6.       The IMF lays the smack down on Brexit.

7.       Finally – someone has actually gone up to everyone in a café and asked them what exactly they’re working on at that moment. Two main thoughts: one, Americans are pretty entrepreneurial; and two, they’re all clearly lying because none of them said ‘goofing off on Facebook’.

 Have a great weekend, everyone!


Posted in Uncategorized | 1 Comment

Intra-household Resource Allocation and Familial Ties

Households in traditional societies often deviate from the form of the nuclear family household that dominate in developed economies. Grandparents and grandchildren, married siblings, other extended family members, or even unrelated individuals may cohabit, produce and consume together.

In sub-Saharan Africa, there is evidence that the prevalence of nuclear family households is rising, in both urban and rural areas. It raises the question what is driving this change and what consequences it may have for the way households allocate resources.

The role of extended families and kinship networks in economic interactions has received considerable attention from economists in recent years (literature reviewed by Cox and Fafchamps 2008) but with a focus on extended family members who inhabit separate households. Whether and how the cohabitation of extended family members affects intra-household allocation is less well-understood.

Agricultural households in Burkina Faso provide an interesting setting for exploring the relation between household structure and decision-making because of the diversity of family ties that exists within the same household. The setting also has two distinctive features which make it easier to formulate and test hypotheses related to household production: (i) the practice of assigning farm plots, individually, to adult household members for which they control production choices, as well as the proceeds of farm output (Udry 1996) and (ii) besides these ‘private’ plots, the presence of ‘common’ plots which are farmed collectively under the management of the household head (Kazianga and Wahhaj 2013; Guirkinger and Platteau 2015).

Using data from a panel survey of agricultural households in Burkina Faso, conducted by the Ministry of Agriculture, we find, within the same geographic, economic and social environment, that nuclear-family households achieve near Pareto efficiency in allocating productive resources and Pareto efficient allocation of consumption, while extended-family households do not.

This pattern is captured in the following figure which shows the distribution of plot yields across farm plots planted to the same crop in the same year (i) belonging to the same nuclear-family household and (ii) belonging to the same extended-family household, after controlling for differences due to the physical characteristics of the plots. For comparison, the figure also shows corresponding distributions for farm plots (iii) within the same village and (iv) managed by the same individual.


For each group, the dispersion of yields captures the potential for improving output by reallocating resources, and therefore a wider dispersion implies greater inefficiency. The household-level distributions, for both subsamples, lie between the village-level and individual-level distributions. This is consistent with the findings by Udry (1996) and Kazianga and Wahhaj (2013) and implies that the household is more efficient than the village at allocating resources across farm plots that belong to the group, but not as efficient as the individual. We also see from the figure that there is greater variation in plot yields across apparently identical plots for extended family households as compared to nuclear family households. The equality of the two distributions is rejected at any conventional level using a Kolmogorov-Smirnov test, implying that the allocation of production resources is more efficient in nuclear family households than for extended family households.

We develop a theory where household members with closer familial ties exhibit higher levels of altruism towards each other (drawing on the evolutionary approach to altruism and familial ties, based on the work of Hamilton (1964)), which in turn motivate them to make intra-household transfers close to that required for efficiency. The theory predicts that household members who share a nuclear family tie (as opposed to an extended-family tie or no family ties) should contribute higher levels of labour on each other’s individually owned farm plots. This is confirmed by data on agricultural labour contributions by different household members on individually owned farm plots.

Using data on consumption expenditures by different household members, we implement two tests of intra-household risk-sharing, based data on consumption expenditures and idiosyncratic shocks to income from specific farm plots, and data on child anthropometrics and shocks to mothers’ farm income. With both approaches we are able to reject the hypothesis of efficient risk-sharing for extended family households but not for nuclear family households.

DC5DXM BURKINA FASO, fair trade and organic cotton project, farmer of cooperative UNPCB in village Kayao

Farmer of cooperative UNPCB in village Kayao (Burkina Faso)

A54F9W Painet hn2128 7607 burkina faso farmers cultivating soil village fada country developing nation less economically developed

Farmers cultivating soil in village Fada      (Burkina Faso)


We argue that the frequent presence of extended-family members and unrelated individuals within the household reflects a household’s response to the absence of markets for labour exchange and risk-sharing: the additional household members provide extra labour (in exchange for room and board and use of household land for farming) and serve as a means of income diversification. Consistent with this argument, we find that household heads with more inherited land and exposed to greater income volatility (due to local rainfall conditions and the characteristics of their inherited land) are more likely to end up with extended-family households.

The two theories of intra-household allocation that have received the most attention in the academic literature and tested most frequently using household data are the Unitary Model and the Collective Model. The Unitary Model, which postulates that the household behaves as if it were a single individual, has been consistently rejected by empirical evidence. On the other hand, tests of the Collective Model, which postulates that household members with conflicting preferences are able to achieve Pareto efficient outcomes, has yielded mixed results — not rejected for labour supply decisions in developed countries or consumption decisions in developing countries, but commonly rejected for production in African households. Our findings regarding the allocation of resources within nuclear and extended-family households provides a way of reconciling these two strands of empirical evidence in the literature that have either failed to reject or have rejected Pareto efficient allocation of household resources.

The wider implication of our analysis is that as markets develop and agricultural land scarcity increases, extended-family households should give way to nuclear family households. This should result in more efficient allocation of resources for production and consumption because of the ties that bind together members of the nuclear family.


Link to the CSAE Working Paper

Harounan Kazianga is an Associate Professor of Economics at Oklahoma State University. His research focuses on rural economic activity in Sub-Saharan Africa. He has conducted extensive research in Africa on technological change in agriculture, the use of financial markets, asset accumulation and gift exchange to cope with risk, gender relations and the structure of household economies, education and a variety of other aspects of rural economic organization.

Zaki Wahhaj is Senior Lecturer in Economics at the University of Kent. His research deals, broadly, with themes of social norms and household decision-making in developing countries, including the role of social norms in intra-household resource allocation and household formation, issues relating to female education and economic participation, and interactions between formal law and customary practices. His current research is based in South Asia and West Africa.

Posted in Uncategorized | Tagged , , , , , , , , | Leave a comment

Links round-up

Hi all,

 So, I’m in Ethiopia (without a mule, sadly), and despite a meeting schedule that could be used as a cudgel (all fascinating, of course, but for someone for whom a blank diary is a form of rapture, also a challenge) I’m having a great time. It’s a fascinating place at a fascinating time, yes; but a lot of what makes it such a great place to be hasn’t changed for a while: kitfo, coffee, and the birdlife. Apparently, so long as one has no shame whatsoever, it’s not entirely frowned upon to bring an enormous bird guide everywhere you go, including meetings. I’ve stopped short at bringing the binoculars because there’s a thin line between eccentric and crazy, and I’ve been known to use it as a jump rope in the past. Anyway. It seems like a come-down after all those amazing endemics, but on to the economics.

 1.       I’m here thinking about, among other things, employment. It’s a complicated and difficult subject, and a good way around that is to outsource one’s thinking to Francis Teal. Here, he considers whether it’s small firms or large firms that policy should promote, and concludes that small firms are good at generating some work, and that work – though badly paid – gets slightly better paid over time, but large firms are where economies become productive. My attempt to square this is thus: it’s the productivity gains, value-added and rents that large firms generate that are crucial to driving up the returns to working in small firms, and so neglecting them will allow you to get lots of people into some kind of work, but limits the rate at which that work becomes better paid. As Francis concludes, “Policy, not for the first time in Africa, seems to be focused on completely the wrong problem.” (Full paper here).

2.       Dietrich Vollrath is the most readable and sensible growth economist out there. In development, we talk a lot about within-sector productivity growth and across-sector productivity growth, terms that are not inherently intuitive. Dietz explains them excellently here, and I encourage everyone to read it all – if the equations scare you, ignore them and read the explanation, which makes everything clear. He’s talking about the US (and briefly South Korea) here, but the intuition applies everywhere, and he summarises it thus: “You cannot talk about aggregate productivity growth without talking about both technology and the distribution of workers across sectors.” In most developing countries, the distribution of workers across sectors isn’t great, and that’s why economists keep banging on about structural transformation.

3.       This week in stuff that makes me so, so angry: social protection and social policy typically evolves based on a strange mix of political expediency, short-term firefighting and ideology. They then persist for a very long time, even when circumstances render them insane. Planet Money looks at US housing support (often awarded as part of a lottery. Yes, a lottery, that’s not a typo), and explains how bad this makes coverage. (Transcript).

4.       Speaking of which, here’s the promised Tim Harford post on the idea of a basic income. a typically intelligent and measured response, he concluding: “the idea appeals to three types of people: those who are comfortable with a dramatic increase in the size of the state, those who are willing to see needy people lose large sums relative to the status quo, and those who can’t add up.” I suspect category three is substantial.

5.       This might be my favourite map of Japan – when you plot the location of electric car charging points, you can no longer see anything else. They might not have the charm of the decrepit and terrifying Ladas that dominate the Ethiopian taxi market, but they’re a hell of a lot greener.

6.       Great LitHub piece on the decisions that underpin what a free market is, and how it looks in different contexts.

7.       Lastly, because I love her, Jessa Crispin burns all her bridges. Every. Last. One. (alternative title: how to be awesome).

 Have a great weekend, everyone!


Posted in Uncategorized | Leave a comment

Should policy seek to promote small firms or large ones in Africa?

Is small beautiful when it comes to firms in poor countries? Whatever one thinks is the answer to that question the pervasiveness of small scale enterprises in countries in sub-Saharan Africa is not in dispute. In a recent CSAE discussion paper, I document that in Ghana not only do small firms dominate the urban industrial landscape but this dominance has been increasing over the period from 1987 to 2003. It thus rather matters if we think this is progress in moving people out of poverty or signs of increasing immiserisation – a rather more prosaic question than the beauty of the small but nonetheless a vitally important question for policy in developing countries.

Newly available data from the Ghana Statistical Office has enabled the pattern of small firm growth to be seen in a longer term perspective and the data is relevant for several issues which have been prominent in discussions of policy in Africa. The reforms in the late 1980s and early 1990s in moving African economies towards more market oriented policies, often under instructions from the IMF, were deeply contentious. How did small firms perform under the new regime relative to the old one and how much did employment in them increase? Small firms are traditionally viewed as part of the informal sector and an influential policy agenda set out by the ILO argues for a focus on ‘decent jobs’. Are the jobs that were created between 1987 and 2003 in Ghana’s manufacturing sector ‘decent’ ones? Finally how successful were these small firms in producing value-added, which is after all what firms are there to do?

Well the answer to the first of these questions is crystal clear. Between 1962 and 1987, roughly the period of the old market hostile regime, the number of small manufacturing firms in urban Ghana scarcely changed. Between 1987 and 2003 their number more than tripled as did employment in them. IMF type policies are good for small firms or, at the very least, do them no harm. And yes small firms created lots of jobs over this period. Data from the household surveys for the period from 1998 to 2005 show that employment in small firms (across all sectors, not simply manufacturing) rose from 3.4 per cent of the population aged 15-65 to 6.7 per cent, an employment rise from 352,401 to 885,391.

But, and it’s an important qualification, these jobs are badly paid relative to those working in larger firms where employment in manufacturing actually fell. So were these ‘bad’ jobs that have been created, ones which the ILO wishes to disappear? Certainly, their report is explicit that a contraction of the informal sector is necessary to promote their agenda However, a rather more relevant question for the poor is – did those working there see an increase in their incomes? Again the answer from the household surveys seems clear. Yes and these income gains were substantial. Between 1998 and 2005 men’s incomes rose in real terms by 45 per cent and women’s by 51 per cent. Poorly paid the jobs were, the key point is that they were becoming better paid.

It is the role of small firms in the development process which is more complex. The Ghana data certainly suggest that their formation is not a problem. But their productivity measured as value-added per employee is low relative to larger firms. Even more important their relative productivity seems to have fallen over the period from 1987 to 2003.

So is small beautiful? The answer appears to be that small firms are good at what small firms do. They create lots of jobs using little capital when policy lets them – in Ghana after 1987 but not before. But large firms are valuable too, they have far more capital per employee and produce far more output – the figure graphically illustrates that while small firm dominate in numbers they produce far less value-added than large ones. So much so that in 2003 the top 1 per cent of firms produced 63 per cent of value-added. Policy rhetoric focuses on the problems faced by small firms. Data from Ghana over the period for which we have it suggests that it is large firms that face the problems. Unloved possibly because they are not seen as beautiful they are vital for the output of the sector. Policy, not for the first time in Africa, seems to be focused on completely the wrong problem.


Figure of the number of manufacturing firms and the total value-added for 1987 and 2003 in urban Ghana.




 bl  br
Note: A small firm employs less than 10, a large firm 10 or more.






Posted in Jobs, Finance and Skills, Uncategorized | Tagged , , , , , , | 3 Comments