Harvard don Dani Rodrik and IMF economist Arvind Subramanian are writing a revisionist paper on the Indian growth experience over the last two decades.
The standard story is that India blundered along the socialist, inward-looking path till 1991. Then came the reforms, which ushered in a new era of growth and productivity. This was based partly on internal liberalisation (abolition of most industrial licensing) but more so on external opening up.
The share of exports and imports in GDP rose stridently, and so did foreign direct investment and portfolio investment. Remittances from overseas Indians (and hence forex reserves) soared because of sensible policies that abolished the black market premium. Opening up culminated in the software-BPO boom that now spearheads India Shining.
The problem with this account, say Rodrik and Subramaniam, is that rapid economic growth actually began a full decade earlier. What’s more, productivity rose phenomenally in the ‘80s, and then either decelerated (agriculture, manufacturing) or accelerated more modestly (services) in the ‘90s.
IMF estimates suggest that annual growth of output per worker increased from 0.86% in the ‘70s to 3.69% in the ‘80s and then decelerated to 3.30 % in the ‘90s. The growth of total factor productivity went from an annual average of minus 2.07% in the ‘70s to 1.28% in the ‘80s and then down to 0.94% in the ‘90s. What accounts for this?
Note that the global scenario was bad in the ‘80s. This decade included the second oil shock, the global debt crisis, the deep recession of 1980-82, and unprecedentedly high real interest rates. So good luck cannot explain the good performance in the ‘80s. Some accounts (Montek Ahluwalia, Little and Joshi) argue that the ‘80s boom was due mainly to an unsustainable government spending spree, financed by foreign borrowing, which eventually led to a bust. However, such Keynesian pump-priming cannot explain why productivity rose so sharply for a full decade.
A third explanation is that the ‘80s witnessed some external liberalisation after the IMF loan of 1981, and an aggressive exchange rate policy after 1985 helped exports to boom. However, Rodrik and Suramaniam show that average import tariffs shot up in the ‘80s, and some easing of non-tariff barriers did not lead to significant changes in the import penetration or import coverage ratios, save for capital goods. India ’s exports /GDP ratio fell to an all-time low in the mid-’80s before rising thereafter. Other explanations include the spread of the green revolution, internal liberalisation through relaxed licensing and MRTP policies, and increased public investment. Possibly they contributed something, but surely not all.
Rodrik and Subramanian feel that a different hypothesis needs to be considered. First, an attitudinal shift by a chastened Indira Gandhi, pushed further by Rajiv Gandhi, changed policies for the better. This attitudinal shift was pro-incumbent rather than pro-competition (broad-banding, automatic licensing, selective delicensing helped incumbents rather than newcomers or consumers). The focus was on internal rather than external liberalisation. The old controls had depressed India so far below its production-possibility limits that even the creeping liberalisation of the ‘80s produced dramatic improvements in productivity (big bang for small buck). Manufacturing played a key role in this, building on spillovers from earlier industrialisation.
The authors conclude that : The transition of the ;80s was as important as that of the \’90s; the ‘80s change was more internal than external; it was more pro-incumbent than pro-competition, more pro-business than pro-consumer. The opposite was true of the ‘90s. While the reforms of 1991 helped India rebound from a crisis, the rebound also owed much to strengths built up in the ‘80s, including spillovers from the earlier import-substitution model.
I think it is true that many observers, especially foreigners, fail to recognise how much was achieved in the ‘80s. Yet the Rodrik-Subramanian thesis needs substantial qualification. An era of sharply rising protection (the ‘80s) typically overstates GDP growth and productivity. High-cost output is shown as high value-added output, and this is economic fiction. In international prices, GDP growth in the ‘80s must have been much slower.
By contrast, a decade of sharply falling protection (the ‘90s) typically understates GDP growth, since the contribution of fictitious value added falls and that of real value added (which is internationally competitive) increases. So, stripped of fiction, GDP growth in the ‘90s must have been distinctly faster than in the ‘80s. Conventional GDP measures show growth in the two decades as being almost the same, but only after the ‘90s were Indian entrepreneurs confident of taking on global competition.
Some call this an improved quality of growth. But I am sure this translated into quantity as well. This is masked by conventional statistics that include fictitious value addition. Here lies a challenge for researchers: estimate non-fiction rates of GDP growth (and productivity growth) in the two decades.
Rodrik-Subramanian are right that the ‘80s changes were more internal than external, and above all attitudinal. Indira Gandhi abandoned the old Garibi Hatao approach, and shifted to creeping liberalisation. Formal policy changes do not capture the full change. In the ‘70s, industrialists were scared of exceeding licenced capacity for fear of penalties or management takeover. But in the ‘80s Dhirubhai Ambani pioneered the technique of constantly creating double the capacity licensed, with impunity. He had many imitators.
Added to this was liberalisation of foreign collaboration. This was external rather than internal liberalisation, and contributed greatly to increased productivity. For example, the auto industry could import no know-how in the ‘70s, and so stagnated. But liberal foreign collaboration in the ‘80s (after Maruti’s entry) laid the basis for gradually creating a world-class industry, in two-wheelers and four-wheelers.
The Rodrik-Subramanian study is not yet finalised. Hopefully, it will take note of my caveats. In my next column I will contest two other conclusions of theirs: that ‘80s reform was more pro-incumbent than pro-competition, and that spillovers from the import substitution model contributed significantly to the boom of the ‘90s.