Prof. Jayanth R. Varma's Financial Markets Blog

Photograph About
Prof. Jayanth R. Varma's Financial Markets Blog, A Blog on Financial Markets and Their Regulation

© Prof. Jayanth R. Varma

Subscribe to a feed
RSS Feed
Atom Feed
RSS Feed (Comments)

Follow on:

Sun Mon Tue Wed Thu Fri Sat

Powered by Blosxom

Wed, 24 Feb 2010

Regulation of mutual funds

Morley and Curtis wrote a very interesting paper earlier this month on the regulation of mutual funds. Their fundamental point is that the open-end mutual fund presents governance problems of a completely different nature from that of normal companies.

Unit holders do not sell their shares in the market, they redeem them from the issuing funds for cash. This uniquely effective form of exit almost completely eliminates the role of voice – investors have no incentives to use voting or any other form of activism.

Morley and Curtis advocate product-style regulation of mutual funds. As I understand this, we must treat unit holders as customers and not as owners of the fund. They also advocate regulations that make it easier for investors to exercise exit rights effectively.

I think this insight is fundamentally correct. In the US context where mutual funds are organized as companies with unit holders as shareholders, this implies a huge change in the regulatory framework.

In the Indian context, mutual funds are organized as trusts and investors are legally the beneficiaries of the trust rather than the owners. Indian regulation already uses exit as a regulatory device. Whenever there is any change in the fundamental characteristics of a fund or in the ownership of the asset manager, the fund is required to provide an opportunity to investors to exit at net asset value without any exit load.

However, the trust structure creates another set of confusing and meaningless legal requirements. The governance is divided between the board of the asset management company and the trustees of the trust. This creates a duplication of functions and regulators might hope that if one of them fails, the other would still operate.

It is however more likely that each level might rely on the other to do the really serious work. The job might be done less effectively than if the locus of regulation is made clearer. There is probably merit in creating a brain-dead trust structure and making the board of the asset management company the primary focus of regulation. This is more consistent with the idea of unit holders as customers.

One implication that Morley and Curtis do not draw from their analysis is that closed-end funds are dramatically different from open-end funds and require totally different regulatory structures. Regulations in most countries tend to regard these two types of funds as minor variants of each other and therefore apply similar regulatory regimes to both. If Morley and Curtis are right, we must treat open-end unit holders as customers and closed-end unit holders as owners.

The governance of a closed-end fund should more closely mimic that of a normal corporation. Regulations should permit the unit holders of a closed-end fund to easily throw out the asset manager or even to wind up the fund. The trust structure in India does not give unit holders formal ownership rights – explicit regulations are required to vest them with such rights.

Posted at 19:58 on Wed, 24 Feb 2010     View/Post Comments (0)     permanent link

Sat, 20 Feb 2010

Short selling and public issues

When the Indian government company NTPC was conducting a public offering of share, it was alleged that many institutional investors short sold NTPC shares on a large scale (by selling stock futures). This gave rise to some talk about suspending futures trading in shares of a company while a public issue is in progress. Thankfully, the government and the regulators did not do anything as foolish as this.

The US has a different approach to the problem – Rule 105 (Regulation M) prohibits the purchase of offering shares by any person who sold short the same securities within five business days before the pricing of the offering. Last month, the SEC brought charges against some hedge funds for violating this rule.

Obviously, Rule 105 is a far better solution than shutting down the whole market, but it is necessary to ask whether even this is necessary. Take for example, the SEC’s argument that:

Short selling ahead of offerings can reduce the proceeds received by public companies and their shareholders by artificially depressing the market price shortly before the company prices its offering.

We can turn this around to say:

Short sale restrictions ahead of offerings can allow companies to sell their shares to the public at inflated prices by artificially increasing the market price shortly before the company prices its offering.

Why do regulators have to assume that issuers of capital are saints and that investors are the sinners in all this? Provided there is complete transparency about short selling (and open interest in the futures market), it is difficult to see why short selling with the full intention to cover in the public issue should depress the issue price.

The empirical evidence is that an equity issue has a negative impact on the share price. This is partly due to the signalling effect of raising equity rather than debt, and partly due to the need to induce a portfolio rebalancing of all investors to accommodate the new shares.

Now imagine that a hedge fund short sells with the intention to buy back in the issue. Since the short seller is committed to buying in the issue, a part of the issue is effectively pre-sold. To this extent, the price impact of an equity issue is reduced. While the short selling could depress prices, this would be offset by the lower price impact of the issue itself.

In short, the short sellers would not change prices at all. What they would do is to advance the effective date of the public issue. If there is a 100 million share issue happening on the 15th and the hedge funds short 20 million shares on the 10th, then somebody has to take the long position of 20 million shares on the 10th itself. For this amount of portfolio rebalancing to happen on the 10th, there has to be a price adjustment and this can be quite visible.

But the flip side is that on the 15th there are only 80 million shares to be bought by long only investors. There is less price adjustment required on that date. The total portfolio adjustment required with or without short selling is the same – 100 million shares. The only question is whether the price adjustment happens on the 15th or earlier.

In an efficient market, the impact of unrestricted short selling would be to force the entire price adjustment to happen on the announcement date itself. The issue itself would then have zero price impact and this would be a good thing.

Because of limited short selling in the past, we are accustomed to issues being priced at a discount to the market price on the pricing date. With unlimited short selling, this would disappear. If the short selling were excessive, the issue may even be priced at a premium as the shorts scramble to cover their positions. It will take some time for market participants to adjust to the new environment. Regulators should just step back and let the market participants learn and adjust.

Posted at 20:30 on Sat, 20 Feb 2010     View/Post Comments (5)     permanent link

Fri, 19 Feb 2010

Taxation of securities

I wrote a column in the Financial Express today about the taxation of securities.

Over a period of time, several distortions have crept into the taxation of investment income in India. The budget later this month provides an opportunity to correct some of these without waiting for the sweeping reforms proposed in the direct taxes code. I would highlight the non-taxation of capital gains on equities, the securities transaction tax (STT) and the taxation of foreign institutional investors.

In the current system, the capital gain on sale of equity shares is not taxable provided the sale takes place on an exchange. Of course, there is STT, but the STT rate is a tiny fraction of the rate that applies to normal capital gains.

The taxation of capital gains is itself very low compared to the taxation of normal income. There is no justification for taxing capital gains at a lower rate than, say, salary income. After all, a substantial part of the salary of skilled workers is a return on the investment in human capital that led to the development of their skills.

Why should returns on human capital be taxed at normal rates and returns on financial capital at concessional rates? Even within financial assets, why should, say, interest income be taxed at normal rates while capital gains are taxed at concessional rates?

The discussion paper on the direct taxes code argues that a concessional rate is warranted because the cumulative capital gains of several years are brought to tax in one year and this would push the tax rate to a higher slab. This factor is in many cases overwhelmed by the huge benefit that arises from deferment of tax.

For example, consider an asset bought for Rs 100 that appreciates at the rate of 10% per annum for 20 years so that it fetches Rs 673 when it is sold. A 30% tax on the capital gain of Rs 573 amounts to Rs 172 and the owner is left with Rs 501 after tax. By contrast, if the owner had received interest at 10% every year and paid taxes at a lower slab of 20% each year, the post-tax return would have been 8%. Compounding 8% over 20 years would leave the investor with only Rs 466. In other words, 20% tax paid each year is a stiffer drag on returns than a 30% tax that can be deferred till the time of sale.

In practice, of course, indexation and the periodic re-basing of the original cost of the asset makes the tax burden on capital gains even lighter. There is no economic or moral justification for subsidising the wealthy in this manner.

Ideally, tax rates should be calibrated in such a manner that equal pre-tax rates of return translate into equal post-tax rates of return regardless of the form in which that return is earned. This might be too much to ask for, but a near zero tax rate for capital gains earned on equity shares makes a mockery of the tax system in the country and should be redressed as soon as possible.

The STT is by its very nature a bad tax because it is unrelated to whether the transaction resulted in a profit or a loss. The real reason for the STT was to make the process of tax collection easier. In this sense, the STT is best regarded as a form of the nefarious system of tax farming that is shunned by modern nation states.

Some attempts are being made to justify the STT as a form of Tobin tax on financial transactions. Without entering into a debate on whether a Tobin tax is good or bad, it should be pointed out that the STT is not a Tobin tax. This is evident from the fact that delivery-based transactions attract STT at rates several times higher than on the presumably more speculative non-delivery-based transactions. The rate on delivery-based transactions is far higher than any reasonable Tobin tax.

A final argument for STT in lieu of capital gains is that foreigners pay lower taxes in India. Many other countries tax foreign portfolio investors at low rates. Indian investors can make portfolio investments in the US (within the limit of $200,000 per annum permitted by RBI). They would not pay income taxes in the US on their income from this investment and would pay only Indian taxes.

There is symmetry here to the US portfolio investor paying taxes in the US but not in India. The only difference is that the foreign investor into India has to come through Mauritius while the portfolio investor into the US can go in directly.

We should also exempt foreign portfolio investors from taxation without forcing them to come via Mauritius. The real problem with the Mauritius loophole is that it allows even non-portfolio investors to avoid Indian taxes, but that is a different topic altogether.

Posted at 14:11 on Fri, 19 Feb 2010     View/Post Comments (0)     permanent link

Thu, 18 Feb 2010

Currency values since the gold standard

I did some analysis of how the value of various currencies have behaved in the hundred years or so since the last decade of the gold standard. I found the results interesting and I am posting them here.

I focus on the twelve countries which are either part of the G7 today, or were one of the eight great powers before World War I, or figure in the top seven traded currencies according to the BIS survey of 2007. I have started with the gold parities during the last few years of the gold standard from 1900-1914. All the currencies of interest were on the gold standard by 1900 and did not change their parities during this period.

I then convert the gold parities into exchange rates against the US dollar. Next, I take into account all the redenominations in which some hyper-inflating currencies had a few zeroes lopped off during the last hundred years. Finally, I take into account the re-denomination of several European currencies into the euro. This leads to the re-denominated gold standard value in USD. I would welcome corrections of any errors that you may find in my data and my computations.

The re-denominated gold standard value tells us what the exchange rate of the modern currency should be to preserve the gold standard value of the old currency through all the redenominations. I compare this with the actual value of the modern currency and compute an annual percentage change in currency value (taking the time period from the gold standard days as a nice round hundred years).

Only two currencies have appreciated against the US dollar over this long period – the Swiss Franc and the Dutch guilder – while the Canadian dollar has held its own. Switzerland has enjoyed a great deal of geo-political luck during this period, but the performance of the Dutch guilder is truly amazing. The euro is commonly regarded as a successor currency of the Deutsch Mark, but if we go back beyond World War II, it makes greater sense to regard it as the successor of the Dutch guilder.

The data has been split into two tables to fit the width of the page better. Countries have been listed in the order of their current GDP.

Country US Japan Germany France UK Italy
Gold Standard Currency dollar yen mark franc pound lira
Grams of gold 1.505 0.752 0.358 0.290 7.322 0.290
Gold standard value in USD 1.000 0.500 0.238 0.193 4.866 0.193
Re-denomination 1.00E+12 x 1.95583 100 x 6.55957 1936.27
Current Currency dollar yen euro euro pound euro
Re-denominated gold standard value in USD 1.0000 0.5000 4.66E+11 126.5684 4.8665 373.6078
Current value in USD(mid Feb 2010) 1.00 0.01 1.36 1.36 1.56 1.36
Annual change 0.00% -3.74% -23.33% -4.43% -1.13% -5.46%

Country Canada Russia Switzerland Australia Netherlands Austria
Gold Standard Currency dollar ruble franc pound guilder krone
Grams of gold 1.505 0.774 0.290 7.322 0.605 0.305
Gold standard value in USD 1.000 0.515 0.193 4.866 0.402 0.203
Re-denomination 5.00E+15 0.5 2.20371 1.00E+4 x 13.7603
Current Currency dollar ruble franc dollar euro euro
Re-denominated gold standard value in USD 1.0000 2.57E+15 0.1930 2.4332 0.8858 2.79E+04
Current value in USD(mid Feb 2010) 0.96 0.03 0.93 0.90 1.36 1.36
Annual change -0.04% -32.22% 1.58% -0.99% 0.43% -9.45%

Not surprisingly, gold itself has done better than any currency, appreciating 4% annually against the US dollar and 2.4% annually against even the Swiss franc. I do not know what the average Swiss interest rate has been during this period; it is conceivable that it compensates for most or even all this depreciation.

What about the Indian rupee? From its gold standard parity of 32 US cents (Rs 15 to the British pound), it has fallen to about 2 US cents – an annual depreciation of 2.67%. This is bad, but better than the Japanese yen and the French franc. The rupee entered the gold standard at a low value reflecting the depreciation of the old silver rupee during the global demonetization of silver in the late nineteenth century. The depreciation of the rupee began only in 1967. The last fifty years would be a lot worse than the last hundred years.

Posted at 18:05 on Thu, 18 Feb 2010     View/Post Comments (1)     permanent link

Sat, 06 Feb 2010

SEC under Schapiro after one year

Mary Schapiro took over as the Chairman of the US SEC a year ago when the SEC’s reputation was in tatters. In a speech yesterday, she reviewed the achievements of the past year.

First of all, she believes that the problems of the SEC were at the top and that if the leadership at the top is changed, the rest of the organization had the competence and attitude required to deal with its challenges. Schapiro says: “having served as Commissioner 20 years earlier, I knew what the agency was capable of ... through this process, I witnessed firsthand the dedication and expertise that I had long believed embodied this agency”

Schapiro believes that the new hires she has made and the new technological initiatives that she has undertaken will be sufficient to enable the SEC to recover its last glory. My own sense is that the problems are not merely at the top but permeate the whole organization. The report on the Madoff investigations revealed problems at all levels of the SEC (see my blog post last year).

Schapiro lists the SEC’s achievements in enforcement during the last year and refers to the case that has been brought against State Street Bank. She says nothing about the Bank of America case where the judge has taken the SEC to task, and subsequently the New York Attorney General has seized the initiative.

On the rule making agenda also, Schapiro is unable to list too much of significance. Even the utterly misguided short sale restrictions of the crisis is cited as an achievement. Some minor changes on proxy rules, rating agencies and money market funds are also cited though none of these changes went far enough to make a serious difference.

Schapiro came to the SEC with high expectations after a succession of lacklustre leaders. I think she needs to do a lot more to fulfill these expectations.

Posted at 22:27 on Sat, 06 Feb 2010     View/Post Comments (0)     permanent link