Mental health plays a powerful role in the decisions we all make every day. However, for someone who is already vulnerable because of a lack of income or their poverty status, psychological well-being can determine the future.

Mo Alloush, MRR Innovation Lab principal investigator and economist at Hamilton College, has been studying the relationships between psychological well-being and poverty. Using data from South Africa, Alloush found that low levels of psychological well-being—specifically depression—can stunt people’s futures.

In this Q&A, Alloush discusses some of the relationships between mental health and income, including the kinds of development programming that can support people struggling with both poverty and depression.

You have spent years studying the relationship between psychological wellbeing and poverty. How did you begin to explore this idea?

To a certain extent it had a non-traditional start.  When I first started thinking about this question, the main source of difficulty was that you have reverse causality with psychological well-being and people’s poverty status. The two are highly correlated and identifying how one influences the other with data is not an easy task.

There were several studies that looked at the effect of changes in income on psychological well-being. You have income shocks that come in many forms, from financial crises to winning the lottery, and people used those to look at impacts on psychological well-being.

It is trickier to figure out the other part, which is how changes in psychological well-being affect economic outcomes, particularly when psychological well-being declines. There are studies that followed people who are suffering from depression and people who sought out treatment. There were a number of randomized controlled trials looking at impacts of therapy and other forms of treatment among specific subsamples.

In my work I was thinking about the feedback loop that could exist if there are causal effects in both directions. I used a unique panel dataset from South Africa, and all four waves of observations had data on psychological well-being, which is pretty rare in this context.

I use dynamic panel data methods to untangle the relationship between psychological well-being and poverty. This approach showed how psychological well-being affects the way people move into and out of poverty. It also showed that people who suffer from or are prone to depression and are poor are especially vulnerable to shocks and have a bigger likelihood of remaining poor.

The low end of my results suggests that all people who are in poverty have a probability of not being poor in ten years. For people who have lower levels of psychological well-being, who are likely suffering from depression, their likelihood of getting out of poverty in five years is 20 percentage points lower.

What does this add to the discussion about resilience when it comes to development programming?

Sometimes we as development economists are surprised when some proportion of people in poverty don’t recover from a shock as quickly as we imagine they should, as quickly as our models suggest they should. Part of my paper’s contribution is that we are not accounting for things like psychological well-being. For people who are particularly vulnerable, an external shock might take a toll on their psychological well-being that makes it more difficult for them to recover.

We’re not talking only about very debilitating psychological episodes. Even mild depression can change the way people react day-to-day. While this won’t affect everyone, for a few it might end up having significant impacts on them. Taking into account psychological well-being is important to understanding resilience. It can also explain why some shocks end up lingering for a long time.

When you talk about depression, what exactly do you mean?

Depressive symptoms are fairly common, and they come in many shapes and forms. One of the things that can characterize depressive symptoms is feeling like it’s harder to get up and go in the morning, or having trouble sleeping at night, not feeling as if you are as good as others or feeling pessimistic about the future.

All those things feature into economic modeling. Our expectations about the future determine our current decision-making. If we’re depressed, our view of what we expect to happen in the future is probably more negative than it actually is, so we might make different decisions. Pessimism is a symptom of depression.

Let’s say you are unemployed, and if you are not having a hard time getting going perhaps you’re actively looking for work. If you are suffering depressive symptoms, you might be a little more reluctant to go down the street and ask if someone is looking to hire. This is not everybody, but on average depressive symptoms can have a huge effect.

What are the implications of this relationship between psychological well-being and people’s livelihoods when it comes to development programming?

Targeting is one. Another idea is to complement programs that give physical assets or cash or training with programs aimed at improving psychological well-being. In 2016, when I first started to draft my paper on depression and income in South Africa, the research in economics in this area was limited. Since then, there has been an explosion of interest in psychological well-being.

There are several development programs that aim to move the needle on psychological well-being, but the results have been mixed. For example, studies have found that program components focused specifically on psychological well-being don’t add much without cash.

One program in particular targeted support for women suffering post-partum depression, and it had significant effects in the long-run three years later. One in particular found that this kind of program for mothers increased their income, the probability of working ad even that mothers who received some psychological care invested more in their children.

A different way to think about targeting is that people who are not suffering from depression will likely end up doing better with a given development program. I would argue against this idea of excluding people who are depressed from programs that might benefit them because people who are suffering depression are likely more vulnerable. Perhaps for those people, in addition to cash they may need more help.

Excluding people suffering depression from programming will also significantly reduce the number of people enrolled in a program. In my sample from South Africa, among the poor it was 30 percent who had an elevated level of depression. In a dataset from Kenya’s pastoralist communities, that proportion of depression among people in poverty was over 40 percent.

The number itself differs, but you find this same pattern around the world. There’s clearly a need for programming that supports psychological well-being.