What Happens to Kids When Their Government Runs Out of Money?

• A study of more than 1.9 million children in 67 countries finds austerity increases the odds of child poverty by an average of 14 percent.
• The two biggest moderating factors are (1) material assets at the family level and (2) government spending on education at the policy level.
• The study uses a unique machine learning method to merge the small-scale processes uncovered by family sociologists and large-scale processes studied by political sociologists.

What is economic austerity?

Austerity is an economic policy that features spending cuts, tax increases, or both. In lower- and middle-income nations, especially, austerity policies can produce a lot of economic hardship.

In their recent article in Social Science Research, Adel Daoud and Fredrik D. Johansson of Chalmers University of Technology in Sweden have attempted arguably the most comprehensive analysis of the effects of austerity on child poverty to date.

Read on to learn more about Daoud and Johansson’s findings, and what you can do to combat the ill effects of economic austerity.

Background

Poverty is damaging in all years of life, but it is especially harmful to children. We know from multiple studies that child poverty is associated with lower educational attainment, lower adult income, worse health, and many other adverse outcomes that follow children through the rest of their lives.

Unsurprisingly, child poverty rates tend to rise when governments implement austerity policies that reduce government spending, increase tax burdens, or a combination of both. That’s not to say that they are always “wrong” or “bad,” as austerity measures at the right time and place can also have economic benefits. But even in these best-case scenarios, large numbers of children suffer long-lasting consequences whenever austerity policies are enacted.

How bad are these consequences? And who is most likely to suffer? These are two of the questions Daoud and Johansson seek to answer in their study. They do so by examining what happened between 1995 and 2005 in the 67 nations where the International Monetary Fund (IMF) was active:

• Founded in 1945, the IMF exists as part of the United Nations. Its goals are to help ensure the stability of the global monetary system and provide assistance to member countries facing economic difficulties.
• It is a “lender of last resort” to nations that cannot generate their own funds and meet their own financial requirements. As national economies teeter on the brink of collapse, the IMF will provide loans to help stabilize the situation.
• It is often a condition of these loans that the nation accepting them must implement austerity measures: raising taxes on their citizens while also reducing government spending. Accepting IMF aid, as a result, nearly always results in a short-term economic shock.

Daoud and Johansson are not just doing a straightforward study here. In fact, they propose to do something incredibly ambitious: to unite the micro-dynamics of households during an economic crisis with the macro-processes that occur within a nation during and after the economic shock. Such a unification, they argue, would be incredibly valuable, especially in its ability to identify the differing effects of economic turmoil across different subgroups.

The Study

The study design is highly technical. But in the end, it is constructed to combine the effects of three factors:

• Families’ own capabilities to protect their children through various resources, financial or otherwise. You might think low-income families have fewer resources and are therefore less likely to be able to weather an economic shock–but some studies suggest middle-class families are just as vulnerable, in part because they have more to lose.
• Demographic and economic factors, such as whether older pensioners must return to work, or where people live within a country (e.g., rural areas may be more affected by cuts than urban areas).
• Political and policy factors, such as the kind of deal a nation is able to work out with the IMF, and how much a government spends on health or education.

The authors were able to use data from the Demographic and Health Survey (DHS) and Multiple Indicator Cluster Survey (MICS), two highly-regarded and nationally representative surveys with high response rates and standardized data collection across multiple countries and years.

• The authors chose 1995-2005 as their sample years to avoid having their data affected by the major health initiatives launched by the Gates Foundation in the mid-2000s.
• In all, their sample covers 1,940,734 children in 567,344 families in 67 countries.
• The authors claim their sample is representative of about 2.8 billion households, or about half the world’s population.

To measure child poverty, they looked at several variables (percent afflicted in parentheses):

• Lack of adequate shelter (51%)
• Lack of adequate sanitation (28%)
• Lack of easy access to drinking water (24%)
• Lack of adequate immunizations or medical treatment (22%)
• Lack of access to radio, television, telephone, and newspapers (18%)
• Evidence of severe malnutrition (15%)
• No formal education (14%)
• Absolute poverty, defined as children suffering from two or more of the above (48%)

The authors chose to use machine learning to strengthen their analysis for the following reasons:

1. Detecting Patterns in Big Data: Machine learning (ML) is excellent at finding patterns and relationships in huge amounts of data that humans might not notice. This means researchers can better understand how different factors, like family income or government policies, affect child poverty across many countries.
2. Handling Complexity: Traditional research methods often assume that relationships between things (like income and health) are straightforward and consistent across different situations. However, real life is more complicated. ML can handle this complexity, identifying how these relationships might change in different contexts or for different groups of people.
3. Predictive Power: ML isn’t just about understanding what’s happening now; it’s also good at predicting future outcomes. This can be incredibly valuable for policymakers who need to know the likely effects of their decisions on child poverty.
4. Improving Over Time: One of the coolest things about ML is that it can learn and improve over time. As more data becomes available, ML models can be updated to provide even more accurate analyses and predictions.

Their methodological contributions to the field are just as valuable as their findings, and can be summarized as follows:

1. New Insights on Poverty: By integrating family and political sociology, this study offers fresh perspectives on how economic shocks, like those from IMF programs, impact child poverty. It uses causal modeling to show that IMF programs can increase the chance of child poverty by 14 percentage points on average, with an even higher increase in some countries.
2. Methodological Innovations: The study introduces a hybrid model that merges predictive and explanatory approaches, enhancing our understanding of social theories and causal relationships. It also demonstrates how integrating computer science methodologies with sociological theory can uncover the varied impacts of economic policies across different populations.
3. Machine Learning’s Role in Sociology: The research identifies crucial factors influencing how IMF programs affect child poverty by applying ML–specifically, the generalized random forest (GRF) algorithm. This approach allows for a more nuanced understanding of policy impacts, potentially informing more targeted and effective interventions.

Findings

The main findings can be summarized as follows:

1. Variation in IMF Program Impacts: Although the average increase in child poverty was 14 percentage points, the study finds significant variability in how IMF programs affect child poverty, with nearly equal contributions from differences between countries and within them. This suggests both broader economic policies and individual family circumstances play crucial roles in determining the impact of economic shocks.
2. Importance of Family Wealth and Government Spending: Surprisingly, middle-class families are found to be as vulnerable to falling into poverty as lower-class families when faced with IMF programs. This challenges the common assumption that economic crises primarily harm the poorest.
3. Implications for Education and Policy: Additionally, the level of government spending on education prior to an IMF program emerges as a critical factor, with higher pre-program spending associated with greater increases in child poverty following IMF interventions. This underscores the need for further investigation into how reductions in education budgets, often a condition of IMF programs, directly impact children’s welfare and future prospects.

What Can You Do?

It may seem as though you, reading this, are too far away and powerless to affect changes to child poverty rates or austerity policies. But here are at least seven actionable steps you can take to make a difference:

1. Stay Informed: Understand the issues at hand by staying updated on economic policies and their impacts on vulnerable populations. Knowledge is power, and being well-informed allows you to contribute to conversations, advocate for change, and support policies protecting children from poverty.
2. Support Local Charities: Many charities work directly with children and families affected by poverty. Donating your time, resources, or money can make a tangible difference in the lives of those struggling to overcome economic hardships.
3. Advocate for Change: Use your voice to advocate for policies that protect and support children and families facing economic shocks. This can include writing to your representatives, participating in advocacy groups, and using social media to raise awareness about safeguarding vulnerable populations during economic crises.
4. Educate Others: Share your knowledge about the effects of austerity and economic shocks on child poverty with your community. By educating others, you can help build a more informed public capable of demanding better policies from their leaders.
5. Volunteer: Look for volunteer opportunities with organizations that provide educational, nutritional, or healthcare support to impoverished children. Your time and effort can contribute to immediate relief for those affected.
6. Support Education Initiatives: Education is a powerful tool against poverty. Supporting initiatives that provide educational resources, tutoring, and scholarships to underprivileged children can help break the cycle of poverty and open up future opportunities.
7. Participate in Local Government: Engage with your local government to support or propose initiatives to reduce child poverty. This can include attending town hall meetings, voting for measures that increase funding for social services, and participating in community planning sessions.

By taking these actions, you can contribute to efforts to reduce child poverty and mitigate the effects of economic shocks and austerity policies. Every effort counts, and together, we can work towards a future where all children have the opportunity to thrive, regardless of their economic background.

Have you or someone you know been affected by austerity policies? Do you have ideas on how we can collectively work to combat child poverty and the adverse effects of economic austerity? Share your thoughts, stories, and suggestions in the comments below.

Stay informed and inspired! Subscribe to our newsletter, “Society This Week,” and get a free copy of our exclusive report, “Peaceful, Powerful: 225 Nonviolent Tactics for Impactful Activism!”