People matter just as much (or perhaps more) in governments as in the private sector. Especially people in leadership positions and critical functional roles. In their search for better outcomes in public policy, governments tend to expend effort and resources on financial support and regulatory measures while overlooking personnel choices. This bias is also reflected in public commentary and academic research that shapes public narratives.
If they are intent on reform and impact, personnel choice decisions are the lowest hanging fruit for governments. In most policy areas, the range between the opportunity cost of a bad personnel choice and that of a capable personnel is perhaps greater than any other policy intervention or reform.
In this context, Philipp Barteska and Jay Euijung Lee have a most impressive new paper which examines the impact of the bureaucratic capabilities (of export promotion officers) on the effectiveness of industrial policy in terms of export performance in South Korea. They use historical data drawn from the three-year rotation of officers (Director) posted in the overseas offices of South Korea’s Trade Promotion Agency (KOTRA) to lead trade promotion activities. After its founding in 1962, KOTRA opened 75 country offices in two decades.
Their headline finding is striking
We exploit the three-yearly rotation of managers of South Korea’s export promotion offices in 87 countries between 1965 and 2000 to show that a one standard deviation increase in bureaucrat ability boosts exports by 37%. Under higher-ability bureaucrats, South Korean exports respond more strongly to a country’s import demand, suggesting a more effective transmission of market information. Higher-ability bureaucrats are also more likely to serve multiple appointments, consistent with first appointments functioning as screening devices for bureaucrats. Lastly, bureaucrat experience matters – the products a bureaucrat is exposed to during their first appointment also see export increases in subsequent appointments. This signifies that while organizational capacity can grow endogenously, it might also exhibit path dependence.
The study uses a context with a clear causal identification pathway.
The bureaucrats we study… manage offices targeting a common outcome variable, defined at a geographic level: exports to the country where the office is located. This setting allows for a unique quantification of the role of bureaucrats in industrial policy implementation during a growth miracle.
They make three important contributions
First, we provide evidence that the effect of an industrial policy on export promotion crucially depends on the ability of the directors of overseas export promotion offices. Second, we find empirical support that the policy’s effect can be increased by screening bureaucrats based on performance in their first appointment. Third, we show that learning-by-doing can build a bureaucrat’s capacity, implying a novel channel for path dependence in organizational capacity.
They provide resounding evidence on the impact of export promotion activities.
We use the offices’ staggered roll-out to estimate the effect of opening an office. Exports increase by 38% in the ten years after an office opening. Our main result demonstrates that the effect of the policy on exports strongly depends on the bureaucrat assigned to manage a country office. We use a movers design in a two-way fixed effects framework… We find that increasing bureaucrat ability by one standard deviation increases exports by 37%. In combination with the estimated effect of an office opening, this suggests the policy of overseas export promotion through these offices would have no effect if implemented by bureaucrats one standard deviation below average…
The large differences between the abilities of office managers are explained by one key mechanism: Better bureaucrats transmit timely market information more effectively. We show that, upon the appointment of a high-ability bureaucrat, exports of a product rise much more responsively to simultaneous growth in the product’s import demand (or export supply)… we show that bureaucrat experience in certain products raises their exports relative to other products… Similar to differences in bureaucrat ability, experience in a product causes exports to increase more strongly when import demand (export supply) goes up… the country fixed effects represent much more than just the overseas offices. The country fixed effects encapsulate time-invariant gravity variables, such as distance and market size, which have been found to be very strong predictors of bilateral trade volume. We find a negative correlation between bureaucrat and country fixed effects, suggesting that better bureaucrats work in smaller countries. Overall, bureaucrat and country fixed effects jointly explain around 90% of the spell-level variation in exports (after removing product-year trends).
The headline finding from the study is the outsized importance of bureaucrats in driving export performance.
Bureaucrats explain a substantial amount of variation in Korean exports… they explain around 14% of the variation. Furthermore… one standard deviation of bureaucrat ability is estimated to be 0.318, implying an increase in the dollar value of exports of 37%. Moreover, the magnitude is comparable to the policy’s average effect – the effect of opening an office – of 0.321 (38%)... Hence, opening an office would not impact exports if the office is headed by a bureaucrat whose ability was one standard deviation below the mean. To put the magnitude into perspective, we compare the effect of a one standard deviation increase in bureaucrat ability to that of geographical distance. Assuming an elasticity of trade to distance of -1, it amounts to roughly the effect of reducing trade distance from London-Seoul (8,900km/5,500 miles) to Mumbai-Seoul (5,600km/3,500 miles)… An alternative way of quantifying the effect size is to consider how much more attractive a KOTRA office makes a country as a destination for South Korean exports. This suggests an office opening makes Ecuador – a country with a fixed effect at the 25th percentile – as attractive as Greece – a country at the 50th percentile. At the same time Greece with an office is as attractive as Spain – a country at the 75th percentile…
Korean ambassadors… have broad diplomatic responsibilities, making export promotion just one, if at all, of many duties. In contrast, KOTRA overseas office directors are solely focused on increasing exports… our decomposition exercise… shows that while ambassadors do explain some variation in exports, they account for a much smaller share of the total variation (4%) than KOTRA directors (14%)… there is a strong drop in exports upon the appointment of an ineffective bureaucrat. However, this drop is only relative to South Korean exports to other countries.
They point to the mechanism of impact as the posting of an effective bureaucrat
We show that upon the switch to a more effective bureaucrat, Korean exports increase more strongly for products that see increasing import demand in a given country-year. They also increase more strongly for products that see increasing export supply to other countries from Korea. Our findings suggest that most – but not all – of the effect of high ability bureaucrats comes from more effectively exploiting market conditions, e.g., by relaying information about destination market demand… Switching to a more effective bureaucrat causes a sharp increase in the elasticity of South Korean exports to market conditions. Losing an effective bureaucrat causes a sharp decrease of similar magnitude…
We find a sharp change in the elasticity of South Korean exports to market conditions in line with the new bureaucrat’s fixed effect and going against the old bureaucrat’s fixed effect. The response of South Korean exports to market conditions increases by around 5 percentage points when the bureaucrat ability increases by one standard deviation. This implies an increase in the reaction of South Korean exports to market conditions by around 20% (from a base of around 25%).
They offer some very useful policy implications relevant to the postings of officials.
Underperformers in 1st Appointment are not Reappointed… The study finds that residualized exports during a bureaucrat’s first appointment, part of their estimated fixed effects, are predictive of bureaucrats’ careers… We find a positive significant effect of residualized exports during a bureaucrat’s first ap- pointment on number of appointments of 0.240 (standard error: 0.112)… We find that a bureaucrat’s first appointment typically is to less important countries, while third appointments are to the most significant markets. The opening year of a KOTRA office and a country’s fixed effect serve as proxy for a country’s importance… the organizations experiments with bureaucrats by initially appointing them as managers of less important country offices. While these offices may contribute little to aggregate Korean exports, observing bureaucrats as managers of these offices allows the organization to acquire information about managers’ abilities. An appointment as manager of a small country office plausibly is much more informative about a bureaucrat’s ability than a subordinate role in a more important country.
The paper points to some very interesting counterfactuals that serve as policy insights.
First, Korean exports would rise by 6.8% if single-appointment bureaucrats were as good as multi-appointment ones. This is a relevant counterfactual as KOTRA could operate an additional set of offices in smaller countries to experiment with bureaucrats prior to appointing them to the countries in our data. The estimated difference in ability implies a 25%51 increase in Korean exports from replacing a single-appointment bureaucrats with the average multi-appointment bureaucrat. 30% of appointments are of single-appointment bureaucrats, increasing exports by 6.8% averaging across all appointments.
Second, Korean exports would decrease by 7.1% if ability did not predict a bureaucrat’s number of appointments, i.e., if there was no selection on ability. Replacing a single-appointment bureaucrat with the average across all bureaucrats increases exports by 11%. Doing the same for a multi-appointment bureaucrat decreases exports by 13%. As there are more appointments of the latter type of bureaucrats, the overall effect is negative. Together, these two counterfactuals suggest that the organization is exploiting just above half of the possible gains from selecting bureaucrats in line with its screening ability. Third, Korean exports would decrease by 16% if all bureaucrats were as negatively selected as the single-appointment bureaucrats.
It also points to the importance of bureaucratic experience.
We find that… the point estimates translates into an increase in exports by 3% in products in which a bureaucrat is experienced relative to those products in which the bureaucrat is not experienced… This is the first evidence regarding learning-by-doing as a channel for increasing bureaucratic capacity… Learning-by-doing in an organization also points to a novel source of path dependence in organizational capacity. A bureaucracy will be most effective at carrying out familiar tasks. Expanding into policy areas in which the bureaucracy has no recent experience builds capacity but is less likely to bring immediate policy success…
Similar to the effects of a higher ability bureaucrat, we show that upon the switch to a bureaucrat who is experienced in product p, South Korean exports increase more strongly if this product sees increasing import demand in a given country-year. They also increase more strongly for products that see increasing export supply to other countries from South Korea. Allowing for this triple interaction makes the estimated main effect of experience much more noisy – suggesting that most of the effect of bureaucrat experience comes from more effectively exploiting market conditions, e.g., by relaying information about destination market demand.
This is a brilliant paper. The authors have painstakingly created a database, analysed the context, identified some theories of change, tested several valuable hypotheses, and done robustness checks on multiple aspects to validate their results.
It would be great if someone could do a similar study of the effectiveness of Economic Counsellor Divisions in Indian Embassies and High Commissions and measure the performances of bureaucrats posted there.
It’s also the sort of paper that has messages of first-order relevance to policymaking. It uses quantitative evidence to show that export promotion offices increase exports; good bureaucrats improve the effectiveness of these offices; the performance in their first posting is a good screening for the career management of bureaucrats; learning-by-doing can improve the expertise of bureaucrats; and experienced bureaucrats are more effective. Most importantly, it brings hard evidence on the magnitude of impact made by officer quality in generating outcomes.
An increase in exports by nearly two-fifth with just one standard deviation increase in bureaucratic capability tells us that the quality of officials might matter more than (or at least as much) fiscal incentives and regulatory changes.
No comments:
Post a Comment