Do public works programmes work?

Can the government just 'invent' jobs when times are hard?

Public works programmes are used around the world to tackle unemployment and poverty by paying people to build roads, dams, irrigation systems and other public infrastructure. But do they actually work?

In this episode of Economics Unpacked, we explore the economics behind public works programmes, sometimes called 'workfare', and what the evidence tells us about their real impact on jobs, wages, and long-term poverty reduction. Drawing on economic research from India and Ethiopia, we speak to economists Manisha Shah and Simon Franklin to discuss why public works programmes can be powerful, but are not a silver bullet.

Economics Unpacked brings cutting-edge economic research to the public, breaking down complex policies with world-leading experts. If you’re interested in economics, public policy, or international development, this series is for you!

Transcript

Can the government just invent jobs when times are hard?

For decades, governments around the world have been spending on infrastructure projects as a way to lower unemployment.

One of the most famous examples of this came after the Great Depression in the US, when the New Deal helped to create millions of jobs. Projects included the Lincoln Tunnel, connecting New York and New Jersey, and the Hoover Dam, both still in use today.

Those in favour of public works programmes, also known as workfare, say it’s a win-win, creating more jobs and better infrastructure. But, is it actually an effective way to reduce poverty? And how can we know its overall impacts?

Welcome to Economics Unpacked, a VoxDev series where we speak to experts and explore the economics behind policies that affect millions of people around the world.

Let’s meet our first expert.

Manisha Shah: Hi, my name is Manisha Shah, and I’m an economist and professor at UC Berkeley. We’re seeing these types of programmes being implemented all over the globe today. The world's largest public works programme is NREGA, the National Rural Employment Guarantee Act in India. It basically enshrines the right to work for rural Indians by guaranteeing them 100 days of employment per calendar year. It was enacted by the Indian government in 2005. In 2024, this programme employed 120 million Indians. This is greater than the population of California and Texas combined. Workers are paid around the minimum wage. Projects typically involve very labour-intensive public infrastructure investments, whether this is work related to irrigation, water conservation, roads, and/or the clearance of land for cultivation. NREGA also has a special provision for women, in that women should receive at least one-third of the jobs.

There’s been lots of research showing the benefits of public works programmes, but do they have a wider impact on communities?

Manisha Shah: While these programmes are designed specifically to help poor people, there’s also reason to believe that there could be really important economy-wide benefits of public works programmes. The programme comes into the community, roads and bridges are built. This may help local businesses connect to the larger markets, and we may observe increased productivity. Another example of this might be irrigation systems get built in rural communities that may then help local farmers increase their yields, again increasing productivity.

If the programme does have economy-wide benefits, it would be particularly good value. But how can economists study this?

Manisha Shah: To understand whether NREGA is having these economy-wide impacts, we're going to need to be able to trace these communities and these economy-wide impacts over time. And the reality is that we just don’t have this type of administrative data, especially since we’re interested in some of the most poor and most remote rural districts in India.

So what did they do without this administrative data?

Manisha Shah: In the absence of reliable administrative data, it turns out that night-time lights picked up by satellites are a valid proxy for measuring this type of economic activity.

Researchers combined this night-time data with data on rural bank deposits from the Reserve Bank of India. And what did they find?

Manisha Shah: So we find consistent and robust evidence that NREGA increases economic activity by around 1–2% per household.

But as with so many policy impacts, overall averages are masking important variations. The benefits are higher for relatively better-off regions and lower for the poorest areas. In other words, the projects being built in poorer areas tend to be less effective at generating jobs and boosting the economy.

Manisha Shah: And we’re not sure exactly why, but we have a few hypotheses. One potential mechanism is implementation quality. The programme impact really seems to vary a lot depending on where you are, who the programme is targeting, how old you might be.

So there are clear positive benefits from the NREGA programme in rural India, even if they are a bit uneven. But what about other contexts, particularly urban settings?

Enter our next expert.

Simon Franklin: I’m Simon Franklin. I’m a development economist from South Africa, and I’ve been working for many years on cities in the developing world, particularly Ethiopia, and how people work and look for jobs in that setting. In Ethiopia, we think the urbanisation rate has doubled over the last 20 or 30 years. So people are moving to cities in large numbers, and that also means that poverty is going to be more and more concentrated in cities as well.

So, while public works programmes have been shown to have an impact in rural areas, to have a lasting impact on poverty for the largest number of people, they need to work in cities too.

Simon Franklin: The Urban Safety Nets Program, the urban PSNP, is a new programme. It started in 2016 by the government of Ethiopia. This was providing guaranteed work to households that were selected for the programme, to work for 240 days per year per household. And this work was done very locally, community-based work. You could think about cleaning streets, greening spaces, cleaning up sewage areas, public toilets, that kind of thing.

And how did they study it?

Simon Franklin: We studied this programme in its first year that it was being rolled out across Addis Ababa, the capital of Ethiopia. The programme was rolled out into some areas of the city but not others, while the government was testing the model in the first instance. So, what we did was we designed a randomised controlled trial so we could compare whole neighbourhoods that got the programme to neighbourhoods that didn't get the programme. Crucially, we were able to look at the direct effects on people actually doing the work, but also on people who were not in the programme but might be affected by the spillover effects.

The thinking is simple: if public works pull workers out of private work, reducing the supply of labour in the market, therefore push up wages in the private sector. So what did they find?

Simon Franklin: So the big finding of the study is that the programme increases wages in the private sector for low-skilled workers by about 19%, when the programme is rolled out to the whole city. And that was the big surprise was quite how large those effects were. We also found that the programme improved local living conditions. These were noticeable differences to the quality of life, to the quality of public infrastructure and living standards, that were noticed not just by people doing the work themselves, but by their neighbours. Then, of course, the programme also increases the earnings of the people doing the work. And when you combine all of those things together, we estimate very large effects of the programme on the welfare of the urban poor – not just to the direct beneficiaries, but to people who were affected by spillovers.

Now, when policymakers choose a programme, they are often selecting from a range of options, so we should compare against other alternative policies that were also available. For example, what if people had just been given cash instead?

Simon Franklin: And we estimate that those effects were larger than if you had given the equivalent amount of money in cash. Because the cash transfers wouldn’t have had the same labour market effects that the public works do, and they wouldn't have improved the local infrastructure and amenities in the way that public works do.

So, is it really a win-win?

Simon Franklin: Our results suggest that public works can be a very effective way for reducing poverty in urban areas. One surprising finding of our study was that public works didn’t reduce unemployment, and in particular didn't reduce unemployment for the youth. And the government has subsequently developed new programmes to try and deal with this unemployment amongst the youth separately.

And what about long-run impacts on welfare?

Simon Franklin: So we don’t yet have evidence that these programmes have long-run effects, that they would have sustained impacts on poverty. And until we have that evidence, we have to consider that these are short-term social protection programmes rather than fixes to the urban poverty problem.

Now, this doesn’t mean public works programmes do not have long-run effects – just that we don't have data over a long enough period to answer that question.

To wrap up, despite differences between India and Ethiopia’s public works programmes, researchers found two important impacts in both settings. One, they can substantially improve the well-being of the poor. And two, they can have positive spillover effects on the economy. But, they are not necessarily a one-size-fits-all solution.

Manisha Shah: So, I want to note that governments have choices about which types of anti-poverty programmes to implement. India has gone big on public works. Other low- and middle-income countries have gone big on unconditional and conditional cash transfer programmes.

So, decisions should be made on a case-by-case basis.

Manisha Shah: And I think that's where economists come in to use the data that we have to further study these various types of anti-poverty programmes to fully understand which of these have the biggest bang for the buck. Because the ultimate goal is to decrease poverty in the long run.

To keep up to date with these important debates in economics, like, follow, and subscribe. You’ve been watching Economics Unpacked, brought to you by VoxDev.