When is a policy U-turn acceptable?

The Reserve Bank of India recently delivered a small but revealing lesson in economic policymaking. On 5 August, RBI governor Sanjay Malhotra was asked whether the special FCNR(B) scheme, designed to attract foreign currency deposits from non-resident Indians, might be closed before its scheduled 30 September deadline. He said there was no proposal to do so. Nine days later, the RBI abruptly advanced the closing date to 31 August.

There may have been perfectly valid economic reasons. The scheme had worked almost too well. Dollar inflows were strong, while the subsidy cost of providing assured protection against depreciation was mounting. There was the growing headache of liquidity management and future forex liabilities. The issue is not whether the RBI is entitled to change course but why give a categorical assurance to the contrary when the possibility of an early exit is already being discussed.

Economic policy is full of U-turns. Some are evidence of agility and learning. Others betray poor preparation, political pressure or plain unpredictability. How do we tell the difference?

Donald Trump’s tariff policy is the most glaring example of the latter. Since the ‘Liberation Day’ tariff announcements of April 2025, duties have been imposed, paused, modified, lowered, threatened again and sometimes re-imposed. Firms managing supply chains have to brace for a tariff now and for uncertainty over the tariff regime six months down the line. Policy volatility itself becomes a tax on investment.

India has its own examples of U-turns. The three farm reform laws, for example — enacted with an iron hand and then repealed after a determined pushback from farmers. The ethanol blending programme — rushed towards E20 as the default option and then backtracking in the face of mounting evidence of its effect on vehicles and other concerns.

Or take pension reforms. The New Pension Scheme (NPS) moved government pensions away from an unfunded, defined-benefit promise towards a funded, contribution-based system. Political pressure for restoration of the Old Pension Scheme then produced the Unified Pension Scheme, which restores an assured pension element while retaining contributions. Yet by March 2026, only about 1.24 lakh of 24.14 lakh eligible central government employees had opted for UPS.

All these reversals are not, however, the same kind: sometimes the politics changes; sometimes evidence changes; sometimes the original policy was not thought through. Sometimes the policy question itself is badly framed.

Consider the question: ‘Does industrial policy work?’ There can hardly be a universal yes or no. Which industry? What instrument? For how long? Under what governance arrangements? Is support conditional on performance? Is there competition or protection of incumbents? Is India’s Production-Linked Incentive (PLI) Scheme uniformly good policy?

Economics cannot provide watertight answers to policy questions because economic behaviour is not governed by fixed parameters. A policy changes behaviour and changed behaviour alters the environment in which the policy operates. A farm reform changes incentives but also political mobilisation. A production subsidy can prop up an industry, but that industry then becomes a constituency that lobbies for continuation of the subsidy. Political economy is not an external disturbance; it is part of the system.

Even the most hallowed international institutions have had to learn this lesson.

During the Asian financial crisis of 1997-98, IMF-supported programmes in Thailand, Indonesia and Korea initially emphasised tight macroeconomic policies alongside sweeping structural reforms. Malaysia did not enter an IMF programme. It eventually imposed capital controls, pegged the ringgit and subsequently staged a strong recovery. Was IMF policy wrong? Malaysia became an important counter-example to the then prevailing IMF hostility towards capital controls.

The euro-area sovereign debt crisis of 2011 produced another awkward episode. Greek debt was judged to be unsustainable. But the IMF changed its own exceptional-access framework. It did not impose immediate debt restructuring but gave generous aid to Greece. This was because European institutions and governments were firmly opposed to restructuring Greek sovereign debt. The contrast with the rigidity shown towards Asian borrowers in 1997 inevitably led to accusations of double standards. Or was this a U-turn due to fresh thinking about fiscal austerity?

The World Bank provides an even more striking example. Its influential 1993 ‘East Asian Miracle’ report concluded that promotion of specific industries had generally not worked. At the time, it helped entrench the Washington Consensus that market-friendly policies were the way to go rather than state-led industrial planning alternatives. Three decades later, its 2026 report says industrial policy is back and offers governments an evidence-based toolkit for using it. That is good news for India’s PLI but is it not a U-turn?

For years the World Bank published its Doing Business rankings, which acquired enormous influence over how ‘business friendly’ a country was thought to be. Yet ironically, in Doing Business 2010, India was #133, Brazil #129 and Russia #120 out of 183 economies; China was #89. Clearly the World Bank indicators had failed to explain why these poorly ranked BRIC economies were growing so rapidly. Subsequently, following a mini scandal, the Bank in 2021 discontinued its Doing Business rankings. It has been replaced by another metric.

None of this is to say that economists or policymakers cannot or should not change their minds. To stick to your guns in the face of contrary evidence is not consistency, it’s dogma. The problem arises when institutions speak with greater certainty than the underlying economics warrants or when reversals driven primarily by political expediency are presented as economic necessity.

The RBI could have said the FCNR(B) window will remain open until 30 September or till inflows crossed a specified threshold or concerns arose about financial stability. Indeed, one critique of the episode makes precisely this point: pre-specified triggers could have preserved flexibility without sacrificing predictability.

Good policymaking requires humility, transparency, accountability but also decisiveness. Humility means accepting that forecasts and models can be wrong. Transparency means explaining what changed. Accountability means acknowledging the consequences of a reversal. Decisiveness means uncertainty cannot become an excuse for paralysis.

U-turns are not necessarily signs of weakness. Sometimes they show that policy is learning and is agile. But there is a difference between changing direction because the evidence changed and changing direction because the political wind changed. Institutions preserve credibility not by pretending that they are never wrong but by explaining why they changed their minds, what has changed on the ground and why the new course deserves greater confidence.

Ajit Ranade is a noted economist. More of his writing here

Article courtesy: The Billion Press