The economics of Ebola: some new facts

For much of the last two months Tavneet Suri and I have been working with the Sierra Leone team at Innovations for Poverty Action, the World Bank, and Statistics Sierra Leone on a major household survey to get some facts about how households are faring during the Ebola crisis. The results have just been released.

Given the transport disruptions and the threat of infection, the survey was conducted by cell phone, which is not ideal in a country as poor as Sierra Leone. However, this survey used as its base those who had been interviewed as part of the Labor Force Survey (LFS) undertaken in July/August 2014. Thus we had up-to-date phone numbers for 2,764 people (66 percent of the LFS sample). We also had data that for most people was pre-Ebola. Thanks to the hard work of IPA and SSL enumerators, 70 percent of these were reached during the middle of November (meaning 46 percent of the overall LFS sample was reached). Coverage was good in urban areas but weaker in rural areas. The rural results therefore need to be treated with some caution, although the panel nature of the data helps and the numbers reached in rural areas still far exceeds most data collection exercises in Sierra Leone through this period. The response rate is much higher than the World Bank survey in Liberia.

We also released results from the last two rounds of our survey of markets in Sierra Leone, which provides evidence that is very consistent with the results from the household survey.

Of course we don’t have any good identification to distinguish the impacts of the Ebola crisis from other shocks that have hit the economy (like the unseasonably heavy rains or the fall in the price of iron ore). All we show are changes since August. However, it is important to remember that the Sierra Leone economy was growing strongly before the onset of the crisis. Overall, it looks as though the urban informal sector has suffered most since August. This is perhaps unsurprising as this is the sector that would bear the brunt of restrictions on transport, markets, restaurants, and bars. They are also the sector that is most vulnerable to cutbacks in discretionary expenditure which comes with uncertainty.

The percentage of household heads who reported working in the last week in urban areas fell from 75 percent to 67 percent. The percentage working stayed unchanged in rural areas. Hours worked, for those in work, fell everywhere except Freetown.

Revenues from nonfarm household enterprises fell by 40 percent. While 4 percent of nonfarm household enterprises were reported as not operating in August, this rose to 12 percent in November. A third of households reported (in November) that Ebola was the reason that the HH enterprise was not operating.

Roughly half of households still have rice that is yet to be harvested but the main reason given is that it is still raining. Only three households mentioned Ebola as a reason rice remained unharvested.  There were reports of labor shortages but these were mainly at the household level (14 percent of respondents), while only 6 percent reported lack of labor in the community as a constraint. Over half of households reported hiring outside labor despite worries that fear of infection would reduce this practice.

Just over 70 percent of households reported taking some action to combat food insecurity in the week prior to the survey. As we do not have similar data at this point in the season it is hard to say how much of this is due to Ebola.

Our markets data continues to show that food prices are similar to those seen in previous years. The one exception is that imported rice prices have fallen much more than is typical in cordoned areas. Our market price data are consistent with prices reported by households in the cell phone survey.

Our market trader data looks somewhat more encouraging than in previous rounds. This is partly because we have changed our base year from 2012 to 2011 (the previous round of the market survey stopped in October 2012). There was also an uptick in rice traders as the delayed harvest came in. Traders of palm oil and processed cassava (gari) are still below previous years, although the gap closed somewhat in December.