The Resilient Edition
Editor's Note: Of particular note, this Tuesday (September 15th) at 10am Eastern there is a special edition of faiVLive in Spanish covering Digital Financial Services in Latin America. I'll be hosting with Gabriela Zapata moderating, and Kiki DelValle, Barbara Magnoni, and Xavier Faz will be joining us. Register here.
I apologize in advance if the final links on resilience undermine your resilience at the beginning of the week.
–Tim Ogden
1. MSMEs:
I'm going to be spending a lot of my time over the next couple of years thinking and writing about small firms—more on that soon—so expect to see SME-related research and content appearing in the faiV more often. Of course, the definitions of "SME" or "MSME" are so loose and so varied, that saying that something is "about SMEs" doesn't convey much useful information. In many official definitions "micro" goes from 1 to 10 employees, and "medium" can go as high as 500 employees. So here's a quick vocabulary guide to how I'm going to use terms in the faiV.
I come from the microfinance world, where there is a broadly shared definition of microenterprise: almost certainly informal firms with no non-family employees. And to be clear, owners do not count as employees. Any firm that has hired an employee from outside the family is not a microenterprise from my perspective, because the step from no employees to 1 employee is much bigger than the step between 1 to 2 employees and certainly much bigger than any employee threshold thereafter. For me, then "small" is the range from hiring a non-family employee to hiring a professional manager (a person whose primary job is managing other employees). I theorize that's around 20 employees. I don't have a strong prior about where to draw the "medium" line, so I'm going to go with the round number of 100. To summarize, Micro=0, Small=1-20, Medium=20-100. Personally, I'm most interested in small firms.
With that lengthy background let's get to some actual content. CFI is conducting MSME surveys in several countries and has posted data from a first round of surveys in Lagos, Nigeria (summary) and Indonesia (summary). Statistical Service Ghana has conducted a similar survey. No big surprises emerge--big hits to revenue, 10% to 30% of businesses are closing, and many of those that aren't closing are cutting wages, hours or employees (though perhaps less than I would have expected). How to make sense of this data? Here's an Economics Observatory post from Pierre Bachas, Anne Brockmeyer, and Camille Semelet pulling together more data and running some simulations on what lockdowns might mean for profitability, exit rates and payrolls with particular focus on 10 countries, including Ethiopia, Guatemala and Uganda among others.
A lot of my questions about small firms are about labor markets and employees, given how much we've learned about the interactions of labor markets and microfinance and TUP programs. Which allows me to tie back to the last edition and discussion of monopsony in American labor markets. Bassier, Dube and Naidu have a new paper on the theme that came out last week (Twitter thread from Dube here). They look at what happens to the wages of people who move from one firm to another and find monopsony power is pervasive, but especially for "low-wage, high turnover sectors." That feels like it might be relevant to the MSME labor markets in developing countries. Understanding the relative power of workers and owners in small firms is key to understanding whether, how much and how household incomes can be raised through the small firm employment channel. On a related note, here's a new working paper Josh Merfeld on the labor choices of smallholders and non-smallholders in India. Again it's a labor market issue: "consistent with smallholders lacking sufficient wage employment opportunities," they are unable to reallocate their efforts as crop prices change, and reduce their income by about a tenth.
2. Microfinance:
Jonathan and I, along with Simon Quinn, Muhammad Meki and Farah Said have a new piece up at the Economics Observatory on how the pandemic is going to affect the global microfinance industry. A lot of it is speculative—as it necessarily has to be at this point. The big question is, now that moratoria are ending, what are repayment rates going to look like. The group model still plays a big role in microfinance, not through joint liability, but simply as a method to control costs. But groups can backfire when repayments are difficult since it is obvious who is not repaying. Recall this old paper from Xavi Giné, and this one from Emily Breza on peer effects in repayment behavior.
CGAP's Pulse survey is showing that about 30% of loans have been restructured (under moratoria or otherwise). Put that on top of non-performing loans and you're easily approaching about half the loan portfolio. How will MFIs react to the uncertain situation as collections become possible again. Aggressive collections by loan officers were already a concern of leaders of Pakistani MFIs that we interviewed as part of our earlier paper. Here's MFIN's post-moratorium guidance for Indian MFIs, which seems to just basically be re-iterating standards of conduct for loan officers, so it's not just a worry in Pakistan. Here's a new update from M-CRIL on "The liquidity challenge for Pakistan's Microfinance Banks." If you're tracking news about how MFIs are reacting to the end of moratoria and what is happening with repayment rates, let me know.
3. Migration and Remittances:
The remittance mystery continues. If you missed it in the last edition, the mystery is that despite predictions and intuitions that remittances would plummet because of the pandemic, lockdowns, closed borders (and in some cases deportations), in many cases it appears that remittances are growing. The Garment Worker Diaries has two new posts focused on the mystery: an overview of migration and remittances among the workers, and some more in-depth analysis of remittance behavior. This week, I had a conversation with a reporter trying to understand the mystery. He told me that remittance sending companies in Africa are seeing increases in volume and senders, but that right now they attribute most of that to shifts from informal to formal methods of remittance sending, a perpetual challenge in accurately measuring remittance flows.
There are darker stories out there as well, more along the lines of what led me to be very concerned about what might happen to migrants and remittances due to the pandemic. Here's a very disturbing story about African migrants in Saudi Arabia being detained in concentration camps. Reporters smuggled cell phones to them and the pictures they have received are harrowing. And here's the story of migrants on Lesbos, living without shelter after a fire destroyed Europe's largest refugee camp.
There's also new research reaffirming the benefits of migration. Peri, et al. look at the impact of a large surge of Puerto Rican migrants to Orlando after Hurricane Maria. They find that employment and business formation increased, with no decline in the wages of prior residents (though average wages fell slightly since the migrants wages were lower than average). As Lyman Stone puts it, "No evidence a migrant shock reduced native wages. Evidence of a migrant shock increased economic growth. Migration is good folks!" Now somebody let the Saudis, and the Europeans, and the Americans know.
4. Our Algorithmic Overlords:
It's been a long time since we've checked in on our overlords. Here's a story that confirms all of the worst fears of the worst of technology and of humanity coming together: Pasco County has been "predicting" who will commit crimes in the future and proactively having police officers harass those people, their families and their employers(!). You might think I'm making this up, but I'm not. I wrote several years ago, I think, about John Hancock, a large US life insurer, deciding that they would no longer offer policies to people who didn't agree to share information about their activities. From an economic perspective it's hard to argue with the logic--sharing this information deals with the two major frictions of insurance markets, asymmetric information and moral hazard. But it does feel a little different when you read about John Hancock partnering with Amazon to offer "free" tracking wristbands to customers.
Here's a new paper about hiring algorithms and the trade-off between the "exploitation" and "exploration" margins of evaluating applicants. I have to confess I find this topic particularly interesting, because of my fear that I will never be able to get a job in an algorithmic hiring future, because my resume is so odd. Anyway, the paper finds that it is possible to build an algorithm that values "exploration" and that algorithm improves both quality and diversity.
5. Resilience:
There are a lot of people talking about "resilience" in recent years. For instance, here's a story about what lockdowns can reveal about entrepreneurs' resilience in developing countries. CGAP's updated theory of change for financial services prominently features resilience as a one of two key outcomes. Here's a whole spotlight from the New York Times devoted to resilience.
Who doesn't like resilience? Well, perhaps it's overrated. And especially by Americans.
