Counting Everything That Counts

November 14, 2019

Good morning and welcome to the 7th annual IMF Statistical Forum.

The theme for this year’s conference is how to measure the informal economy.

This theme reminds me of a quote, often attributed to Albert Einstein: “Everything that can be counted does not necessarily count; everything that counts cannot necessarily be counted.”

So, with this year’s conference, one could say we are trying to prove Einstein wrong and find ways to count all the things that count, but are not currently counted!

Why measure informal activity

The last two Statistical Forums were about diagnosing and providing recommendations on the impact of digitalization on the accuracy of macroeconomic and welfare indicators. This year’s conference returns to the basics. Needless to say, it is not an easy task. For one, some activities are informal precisely to stay in the shadows, away from tax obligations or laws. But capturing informal activities is not just for tax collection or law enforcement purposes. The informal economy is so much more, and is only growing with the rise of the shared economy. In some low-income countries, for example, the informal sector is estimated to be the same size as the formal one, and perhaps even larger.

If only we knew for sure.

As management guru Peter Drucker said, “if you can’t measure it, you can’t improve it.” It would be an improvement, for example, to assess a country’s debt sustainability, or apply a fiscal rule, if we could measure all economic activity.

And to the extent we know, we can understand economic trends better. For example, when I was working on Poland in the early ‘90s, we documented how informal markets were emerging as substitutes for state sector production. That explained how the seeming decline in food processing was simply a shift to informal production that was not picked up in the official data.

The path forward

Measuring informality is not exactly a new problem. The IMF first established a conceptual framework to measure the informal economy in 1982, with a book called The Underground Economy in the United States and Abroad.

Since then we have made big strides, but more is needed.

In some cases, the answer lies in increasing a country’s capacity to measure GDP. Our Statistics Department has enormous demand for Technical Assistance to improve GDP measurement. By covering economic activity that was previously not- or ill-measured, one country in Africa saw its GDP increase by 89 percent!

Another route is to create incentives for formalization. This requires reforms in various areas, for example taxation, labor law, product markets, and improving the business climate. Countries like Egypt, India, and Mexico have achieved important results with that approach.

Our staff is always exploring new methodologies. We formed a task force on the informal economy in 2017 to identify data collection techniques and compilation methods to reflect the informal economy in external sector statistics. We are working with other international agencies on illicit financial flows. One of the cases you will hear about tomorrow argues that tracking illegal activity in the US national accounts would have raised 2017’s GDP by 1 percent.

Other international organizations are working in the same direction within their expertise. The World Bank, the International Labor Organization, and the UN have been doing excellent work on measuring informality through the labor market, and we collaborate closely with them.

The power of Big Data

A new and exciting area is about using innovation and technology to help with this major challenge. How can big data and AI help measure the informal sector? What innovative sources of data can be leveraged to provide new insights on economic activity? And ultimately, how can developing economies leap-frog using these new techniques?

Let me give you some examples from the upcoming presentations:

Experts from the Chinese National Bureau of Statistics will talk about how transactions on digital platforms can be used to observe informal activities. Since observing is not the same as measuring, one challenge they have is disentangling what is informal from what is already formal.

Researchers from the Central Bank of Mexico will explain how they use satellite nightlights imagery to estimate that the “non-registered” economy in the country could represent up to 29 percent of GDP. This is a technique that Fund researchers have also used to gauge economic activity in Iran and Venezuela, among others.

You will also hear about how the Fund is using big data to estimate economic activity in countries where official GDP and trade data are either imprecise or too delayed, as in the case of Zimbabwe.


To sum up, this year’s conference is essentially a return to core statistical business: measuring economic activity accurately, using all possible means. I hope your discussions will result in practical advice on measuring informality, which is so important and yet so difficult.

I would like to thank you all for coming to our Forum—which, despite all of what I just said, we hope to keep very informal. Let me also thank Louis Marc and his team in the Statistics Department for putting together yet again a great conference.

And now, it is my pleasure to turn it over to our Managing Director Kristalina Georgieva. Kristalina is traveling, but this topic is close to her heart, so she wanted to participate with a recorded statement. Please join me in welcoming her onto the screen.

Thank you.

IMF Communications Department


Phone: +1 202 623-7100Email: