Measuring the Digital Economy
Gross Domestic Product (GDP), the de facto metric of economic health, is in dire need of an overhaul. GDP was created to measure monetary value of all final goods and services produced within a country. This cardinal number is now widely used as a proxy for economic well-being. However, Simon Kuznets, who led the team which formalised GDP, warned that
“the welfare of a nation can scarcely be inferred from a measure of national income.”
Some known areas where GDP fails to account for economic well-being are:
Negative externalities created due to economic growth; like pollution.
Non-market activities like household work.
The 21st century challenge:
Increasingly, as zero or low priced digital services have become ubiquitous (think search, email, social media, with almost 6 hours and 42 minutes spent online per day per capita basis), there is a need to come up with a new way of measuring economic well-being. As per traditional macroeconomic theory, GDP is typically calculated basis what consumers spend for goods/services and since prices of most widely consumed digital services are near zero, they go unaccounted for in GDP calculations. Even for consumer subscription services like Netflix, Hulu, the subscription fees has been found to be a gross misrepresentation of the actual consumer value. A data point which drives this point home is that the share of IT in the US GDP has remained between 4-5% for the last 40 years.
The impact:
Most policy makers rely on GDP data to make decisions across a breadth of dimensions - for fiscal and monetary decisions, to design policy that affect firms offering both digital and non digital goods and services. Since, the digital services are highly undervalued, the resulting decisions and policies are made with skewed assumptions. As economist Robert Solow famously commented in 1987 -
"the computer age was everywhere except for the productivity statistics"
(also called the Solow Paradox), alluding to US's anaemic productivity growth then, which was regained soon in the 1990s productivity boom (with semiconductor and manufacturing innovations). But as a lot of policy makers call for regulation of big tech and design incentives for digital service based businesses, the first step is to understand the nature of and the value being generated from digital offerings.
The solution:
Thus, there is a need to compute economic well-being via value generated in the economy and not pure play consumer spend. Free market economics proposes prices equal marginal cost, but marginal cost for digital services tend to be zero or near zero. Popularised by the economist Alfred Marshall, economics already does provide a base measure to capture economic value: consumer surplus. It is defined as the difference between the maximum a consumer would be willing to pay for the good or service and its price. MIT economist Erik Brynjolfsson and his team has developed “GDP-B” as a concept to capture the economic value of both zero and positive priced digital goods. Their model involves estimating consumer willingness to pay via massive online choice experiments (through both best-worst scaling and single binary discrete choices techniques to obtain ordinal ranking of digital goods, associated monetary values and derive demand curves). For example: asking the survey participants to forgo access to Facebook for a month for $500. This monetary value has been observed to be affected by substitutes (for ex: lesser value attached to losing Facebook access given that there exists Instagram, Snapchat etc), network effects (people with more friends tend to attach a higher value to Facebook) and demographics (for ex: older people tend to attach a higher value to Facebook since perhaps there are higher switching costs for this segment to move to alternative platforms). The benefits of using such survey methods are that:
Scalable to zero or positive priced goods.
Run near real time to measure well-being performance.
And the challenges of using such survey methods are that:
Not as comprehensive and precise as GDP.
Does not capture negative externalities involved with digital consumption like addiction, lost privacy etc.
The "GDP-B" model tends to lie in the middle of the spectrum between precise traditional measures like GDP to more subjective measures like the Happiness Index. With the question of technology regulation heating up globally, including the value generated by the digital economy can help policy makers design rules and regulations to incentivise and subsidise the new innovation streams as well. In the Indian context; players like Gaana (music player integrated with MX player now), Hotstar (video streaming), TikTok (video sharing), Google Pay (UPI based payments app), Newshunt (local news) and others across the digital stack are acquiring users at a blitzkrieg pace and it is imperative for the Indian government to include value generated from digital services in framing digital and content regulations.
