GDP Measurement Has Become A Mugs’ Game; It Is No Longer A Useful Guide To Economic Activity

There has been much debate over whether India’s currently-used GDP data overstates economic activity.

A few months ago, former Chief Economic Advisor Arvind Subramanian stirred a hornets’ nest when he suggested that the new GDP data series may have overestimated GDP growth by as much as 2.5 per cent between 2011-17.

In the latter part of this period, we were reporting above seven per cent GDP growth for several years until the recent slowdown brought the figure down.

In a presentation at Brown University, former Reserve Bank Governor Raghuram Rajan also hinted that our GDP may be overestimated. He believes that “GDP mismeasurement” is one of the factors behind low job-growth and the current slowdown.

The problem with GDP measurement is that it is always prone to mismeasurement given the huge changes taking place in the global and domestic economy.

To the extent that GDP (or gross value added, or GVA) measures only economic activity that has a monetary value, GDP has always been miscalculated. For example, homemakers may make indirect contributions to GDP without earning a single rupee in wages.

But today’s mismeasurements go beyond the usual non-measurement involving unpaid homemakers. With the shift in the global economy towards services, many of which are actually free to the user, accurate measurement of GDP or GVA becomes near impossible.

And even if a monetary value is assigned to unpaid products and services, nobody will think it is of any value since the new number has no income attached to it.

The other reason why GDP measurement is getting tougher is productivity improvement – either in quality or quantity – which may not immediately be visible in the GDP statistics.

As T N Ninan noted in a column in Business Standard some time ago, improved productivity in several sectors may have led to a short-term drop in demand for investment and products, which may be impacting GDP growth right now.

This is visible in many areas.

When trucks are allowed to increase axle loads (as they were earlier this year), and the goods and services tax (GST) reduces turnaround time for vehicles, and good quality national highways improve vehicle speeds, most fleet operators can carry more cargo with the same fleet or even fewer trucks than now. This productivity rise means fewer new trucks will be bought in the short run.

Or take the sharp drop in telecom data and voice tariffs after the entry of Reliance Jio. Since there has been no major growth in total users, and tariffs have dropped even while data usage has soared (India is now the world’s largest per capita data consumer), total telecom revenues have been falling even while services consumed have increased and the quality of bandwidth has risen.

“How is this clear improvement in productivity playing out in the GDP data?” asks Ninan. One can point out that if tariffs have dropped and losses mounted for telecom companies, the GDP could actually have fallen despite a rise in economic activity due to higher productivity.

The efficiency of banking and financial services has also boomed with the rise of digital banking. Today, lending and borrowing can be friction-free, with third-party apps bringing borrowers and lenders on the same platform, reducing the cost of credit, and expanding the potential market for it.

In the short term, these technologies will be damaging the business models of existing banks and financial services companies but the productivity boost in the sector will probably register in GDP numbers only after a lag.

We could also consider the huge sales reported by e-tailers like Amazon and Flipkart, who have grown volumes by sharply dropping prices.

This price drop would have pushed production and sales volumes, but at the level of the actual vendor of products, margins would have thinned, and value-added may have dropped. This too would impact GDP.

In other sectors – real estate, for example – value is being destroyed as the artificial inflation of margins and incomes in this sector has been punctured by reality.

Real estate pricing has been built on the assumption that land is always scarce. Large rents were collected by crooked builders, politicians, and bureaucrats who decide land-use policies and give building permissions.

But economics has caught up with this sector. When prices are too high and rental housing is cheaper, no one will buy houses. Also, land supplies can easily be expanded in costly areas by allowing more vertical building.

Again, one wonders how this value destruction is being captured in GDP data. Also, if available land and properties are being used more efficiently due to the high cost of new properties, how is this productivity gain showing up in GDP?

The same goes for autos. If the availability of taxi-hailing apps is improving the productivity of the asset, how is this showing up on GDP?

An owner-driven car may make only one to-and-fro trip a day, but an Ola or Uber can ferry many passengers many times a day, thus using automobile assets more efficiently and productively.

This would have impacted short-term demand for cars, resulting in a short-term slowdown but the productivity boost is not yet showing up on anyone’s radar.

At another level, consider the sheer number of products – especially services – being given free.

While Gmail, Facebook, Google maps have been free for as long as they have been around, many more services – like news, analysis, blogs and book reviews, for example – have proliferated even as the same services are paid for elsewhere (in newspapers and magazines, and paid book reviews).

A recent article in Financial Times points out that the modern economy is no longer properly measurable using GDP as a tool.

It noted: “GDP has always been a flawed concept. It was, after all, devised in the early years of the 20th century to track industrial output, and cannot capture other aspects of our economic life such as unpaid housework or even some paid services.

“What has really changed is that the explosive growth of technology is pushing more economic activity out of the realm of classic GDP measuring tools. This includes ‘free’ services (say, social media) and huge improvements in the output of technology (such as the rising speed of mobile phones).”

When Alvin and Heidi Toffler invented the term “prosumer” to define a new category of consumer who is often the producer of the same item, they were essentially saying what has always been around: production for personal consumption, as in the case of the unpaid home-maker.

When people produce for free, valuing this production in terms of what people pay for similar things in the monetised part of the economy will bloat the GDP, and yet, when useful services become free, it also means that not much income is being generated by such activities.

At some point, though, they might begin to earn revenues, as Google did with its advertisement inserts served along with “free” search.

One can go on and on suggesting how GDP measurement is going to get more and more wrong as the complexion of the modern economy moves from hard products to soft services, including a growing range of free services but the essential takeout is this: GDP measurement, always a mugs’ game, is going to become an even bigger mugs’ game.

It is time to shift to another metric – paid jobs, part-time or full-time – as the most important measurement of useful economic activity. Let us leave GDP to the experts, who can be guaranteed to get it wrong often.