Ideas & Debate

With technology, GDP figures understate service economy

There is a growing (anecdotal) evidence that
There is a growing (anecdotal) evidence that KNBS gross domestic product figures could be understating the extent of services in the economy. FILE PHOTO | NMG 

A couple of weeks of back, I wrote about the perils of using gross domestic product (GDP) as a measure of household economic wellbeing (which was well received). A key point to note is that GDP, by and large, originated from the post-World War II need to measure the use of material resources in an economy at a time when they were quite constrained.

But at the onset, construction of GDP presented some dilemma to both economists and statisticians. Economist Simon Kuznet, one of the earlier pioneers of GDP, constructed it as a measure of national economic welfare, rather than just that of output.

Essentially, this meant subtracting, from GDP calculations, government spending on armaments and infrastructure.

However, given that, at the time (1930s), wartime spending was beginning to hit crescendo, politicians wanted to demonstrate the positive side to war spending.

Consequently, Kuznet locked horns with other economists, most notably with Milton Gilbert of the US Commerce Department—who argued that government spending was a huge driver of output and could not be ignored. That’s was back then.


There is another dilemma with GDP (that remains unsolved). As economist Diane Coyle states: in an increasingly weightless global economy, a measure of economic output designed for tangible physical production (such as GDP) can be misleading.

The truth is that technological inventions have changed the face of global economy, courtesy of warfare.

Indeed, in the history of humanity, warfare has proved to be the mother of invention. As noted in a number of texts, demands of conflict have given rise to major inventions that characterise modern life.

Think about the computer, the internet, or even antibiotics. In 2005, I bought a 250 megabytes (MB) USB storage disk for Sh1,500. As a university student, it was such a remarkable change to the way I stored my classwork and assignments (having graduated from floppy disks). Yet 15 years later, the same Sh1,500, or at a much lower price, I will get a 32 gigabytes (GB) worth of the same disk.

Consider a Motorolla C113 handset retailing at Sh7,500 in 2003; yet 16 years later, you would get a decent smartphone at a lower price.

Today, you can easily order a ride via the many ride-hailing apps, most notably Uber. There are even brilliant minds trying to disrupt public transport via applications that allow hailing of shuttle as a group-such as Little Cab’s e-shuttles.

On the payments space, I can now deposit money into my bank via my phone-and even pay for my utilities (saving myself several hours of queuing time). You can even book a public service bus to Kakamega via your phone.

Beyond technology, there are other metrics of quality of life improvements. For instance, the percentage of off-grid population in Kenya has declined from 84 percent in 2000 to 44 percent in 2018.

Yet, the modern-day dilemma for statisticians is how to capture those price drops as well as the corresponding increases in quality into national accounts. Does the Kenya National Bureau of Statistics’ calculations of national accounts capture the full extent of these improvements in an increasingly weightless economy such as Kenya’s? I have my doubts.

In 1996, a group of experts, known as the Boskin Commission, investigated this matter in the United States and concluded that by failing to account for the quality changes in an increasingly technologised economy, the US consumer price index had been overstating the rate of inflation by 1.3 percent, and understating real GDP growth.

It’s not surprising. In summary, there is a growing (anecdotal) evidence that KNBS GDP figures could be understating the extent of services in the economy.