The middle income illusion

The middle income illusion

There is no economic evidence to support the notion of middle-income trap.

IT, financial services, banking services, banking services volatile, financial services volatile, banking in IT firms, financial services in IT firms, bfs segment, banking financial services segment, bfs, IT services, indian IT services, bfs firms, latest news, business news, indian express news
The middle-income trap is sometimes defined not with respect to an absolute threshold level of per capita income, but with per capita income expressed as a share of US per capita income.

Is there a middle-income trap? Words have multiple meanings. The word “trap” has three major meanings — trap(1), a situation one can’t get out of (trap shooting comes from that); trap(2), mouth; and trap(3), a horse-drawn carriage. Outside of literature (often old), I have not come across trap(3) being used for a long time, certainly not in India. Trap(2), as in, “shut your trap” is colloquial and also rare in India. There certainly is a middle-income trap, interpreted as trap(2), though people who use the expression mean it in the sense of trap(1).

Economists are usually known for their precision. When they use the expression “middle-income trap”, they should define what they mean by “middle income”, “trap” having hopefully been defined. I have said that economists should be precise. They aren’t always, and presume that like the “Khan Market consensus”, there is a “middle-income trap consensus”.

The World Bank has a definition of middle income. It is a range of per capita income between $996 and $12,055, with $996 to $3,895 defined as lower-middle income and $3,895 to $12,055 defined as upper-middle income (the thresholds are often changed, these are 2019 levels). With a per capita income of around $2,000, India is still a lower-middle income country and $12,055 is a long way off. With a per capita income of $12,055, India’s economy and society will be vastly transformed. Nevertheless, like Vitalstatistix, we must worry about these vital statistics. What happens when we cross that threshold? What happens if the sky falls on our head tomorrow?

These numbers are based on official exchange rates, the so-called nominal per capita GDP or Atlas method figures. But a country’s per capita income is in local currency, that is, rupees. For purposes of cross-country comparisons, they have been converted through a common numeraire, the US dollar. For the sake of sheer perversity, just before 1991 reforms, it was Rs 20 a dollar and right now, it is about Rs 70 a dollar.


Had that exchange rate continued, under an economist’s favourite expression of ceteris paribus(everything else remaining unchanged), the per capita income would have been around $7,000 now — closer to the threshold and perhaps, a reason to worry even more. Typically, when economists use the trap idea, they at PPP (purchasing power parity) dollars, using PPP exchange rates, not official exchange rates. India’s PPP per capita income is now around $7,000. Through sheer coincidence, it is identical to the number obtained by using Rs 20 a dollar as the exchange rate. Do countries get stuck in the middle-income range, PPP or otherwise? Might India get stuck?

The grist to the mill is usually provided by empirical research, documenting the development experience of a diverse range of countries. I know of two such recent surveys of literature on the middle-income trap proposition. The first was by Fernando Gabriel Im and David Rosenblatt, published as a World Bank Policy Paper in 2013. The second was in the Economic Survey 2017-18. Both found no evidence.

The Economic Survey also said, “But, recently doubts about the convergence process have been articulated around the notion of a ‘middle-income trap’. can themselves be traps so it is important to be careful about them.” Who pays attention to such advice when trap(2) beckons? The Survey added: “The reasons for the trap/stall were supposed to be two-fold, operating as a kind of pincer. On the one hand, as countries attained middle-income status, they would be squeezed out of manufacturing and other dynamic sectors by poorer, lower-cost competitors. On the other hand, they would lack the institutional, human, and technological capital to carve out niches higher up the value-added chain. Thus, pushed from below and unable to grasp the top, they would find themselves doomed to, well, middle-income status.

As it turned out, there was neither a middle-income trap nor stall. Middle-income countries as a group continued to grow as fast or faster than the convergence standard demanded.” In this quote, conceptual arguments behind the middle-income trap are also mentioned. The burden of empirical evidence is — there is no such trap. Note two further points. First, a trap cannot be defined without referring to a time-frame. The time series on PPP per capita is a bit more difficult to get than the official rate per capita. With that caveat, take a look at the time series of any relatively more advanced country. Until a few decades ago (a cut-off in 1960 or 1970 will suffice), all these countries were stuck in middle-income traps.

Second, the middle-income trap is sometimes defined not with respect to an absolute threshold level of per capita income, but with per capita income expressed as a share of US per capita income. Even if one uses this relative notion, the case of a middle-income trap existing has not been proven.

Does this mean there are no issues with the Indian economy? Certainly not. After the elections, with a new government in place, plenty of people have come up with agendas for reform. In most instances, these are not short-term quick fixes, but medium-term changes. Therefore, they can rightly be called structural reforms, and the suggestions should be debated, accepted and implemented. Having granted that, the limited point is that nothing is gained by muddying waters through expressions that sound profound, without conveying anything of import.

(The writer is chairman, Economic Advisory Council to the PM. Views are personal)