Yet another harebrained idea

At a recent session on transactions taxes (TTs) on financial securities, derivatives and commodities, I indicated unambiguously that such taxes are a bad idea.

At a recent session on transactions taxes (TTs) on financial securities, derivatives and commodities, I indicated unambiguously that such taxes are a bad idea. I reached this conclusion not in theory but on the basis of research done on experience with TTs around the world. Deputy Chairman, Planning Commission, Montek Singh Ahluwalia, a former colleague — and I hope still a friend — who chaired the session seemed to agree.

HT Image

Such taxes — securities transaction tax (STT) or commodities transaction tax (CTT) — are attractive for finance ministers as an easy way of raising revenue, supposedly painlessly. But they do much damage. They are invisible at first and their benefits are exaggerated by proponents of such taxes. But the impact over time is profound in terms of market impairment and volume migration either off-market or offshore. Even when the tax rate seems small, the cascade effect is very large. And finally, when the damage TTs do becomes visible and obvious, it is very difficult, if not impossible, to reverse.

It amazes me that Indian officials and regulators see risk-hedging as good but assume risk as bad. How does one hedge a risk unless someone is prepared to take it? How does one buy

a future or an option unless someone sells it? So why is the buyer of an option a good risk-manager, but the seller of an option an evil speculator?

Why are these taxes so bad? Let’s look at broad empirical evidence first and logical reasoning later. Since the mid-80s, about 40 countries have introduced STTs in one form or another. Only 14 of them have internationally significant capital markets as far as trading is concerned. Since 1990, eight countries (including Sweden, Denmark, Finland and Japan) have abolished them. That happened despite the fact that in Japan the STT raised about $ 14 billion in revenue (about 4 per cent of total revenues). There is a lesson to be learnt here.

Of the 32 countries that still have some combination of STT, service tax, stamp duty or registration fee adding to transactions costs in their securities markets, 30 have eliminated one or two of these taxes between 1990 and now. All 30 countries have reduced the burden that they have first imposed at least once if not twice. In 15 of the 32 countries with some remaining form of TT, the tax can be avoided by trading off-exchange or offshore. In all these 15 countries, trading volume has shifted offshore. In 12 countries, STT does not apply to foreign investors and non-residents. In China, which last imposed STT and then tripled it to cool its overheated stock market, share prices have collapsed by over 55 per cent (versus an average of 20 per cent for emerging markets overall) and volume has shifted from Shanghai to Hong Kong.

That is correlative evidence, not causal. But what do the several dozen detailed studies done on this subject around the world show? They show that STTs and CTTs:

n Have negative effects on price discovery and market liquidity;

n Have a contra-effect on the volatility of trading volumes, i.e. they actually increase volatility rather than reduce it and also increase price volatility of securities and derivatives;

n Do not just throw sand in the wheels of finance but result in seizing up the dealing engine that drives financial markets, thus reducing market efficiency;

n Adversely affect broker-dealer ability to transform effective demand for securities into realised transactions by impeding inter-dealer trading that is so essential for dealer inventory risk management, thus impairing price discovery, liquidity and market stability;

n Have frictional effects on one asset class that introduce distortions in pricing and price discovery of that and other assets, including derivatives;

n Shift trading volume to other markets/assets not subject to STT.

The studies also suggest that it’s difficult, if not impossible, to design a non-discriminatory STT affecting asset portfolios equally, thus inducing STT-driven portfolio choices. Finally, STT introduction in every country has always had a price-depressing impact at the time of introduction and serial distortion effects thereafter. There is much more that can be said to show why such taxes have effects opposite to those intended and have a number of unintended consequences besides. But that would get tedious.

Let me conclude with the example of what happened in Sweden and let the reader make up his own mind. STT was introduced on stocks and derivatives in Sweden in 1983, doubled in 1986, and abolished in 1991. The revenue impact was minimal. Trading migrated to London when foreign investors avoided the STT. Domestic investors used offshore accounts and foreign brokers.

Despite all this Sweden went ahead and broadened STT to cover bonds as well. That resulted in sharp drops in the trading and prices of Swedish government bills and bonds and in related derivative contracts. This undermined the Bank of Sweden’s ability to conduct monetary policy. Bond trading fell so dramatically that popular support for STT vanished. STT on fixed income was abolished in 1990 and all such taxes were eliminated in 1991. What are the lessons from Sweden’s tryst with these taxes? First, STT can and will be avoided to the extent possible. Second, markets and volumes move offshore. Third, trading shifts to other assets. Fourth, contracts are devised to defeat the STT. Fifth, when STT is removed, some volume migrates back, but there is always a permanent loss to other markets.

After Sweden, seven other countries have abolished STT. Britain is committed to moving toward that goal. So why is India swimming against the tide of sense and experience?

Percy S Mistry is chairman, Oxford International Group