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

Thu, 28 Jul 2005

Are Financial Centres Worthwhile?

The Times of London reported earlier this month that London was the most successful financial services industry in the world with record net exports of £19 billion ($33 billion).

I was struck by the smallness of this number. If this is what the world's premier financial centre can achieve, then is it at all worthwhile for a country to promote itself as a regional or global financial centre? For example, does it make any sense for India to build Mumbai up as a regional financial centre? Perhaps not.

It is reasonable to assume that in a few years time, India‘s software and BPO exports will be above $33 billion. On the other hand, any new financial centre will have a fraction of the net exports of London. Not only will it be smaller, it will most probably not have institutions like Lloyds of London which contribute significantly to those net exports (insurance accounted for a third of London's net exports). We might be better off importing financial services and exporting software!

My colleague, Prof. Sebastian Morris, points out however that the advantage from having a regional financial centre is likely to last for a very very long time. If London‘s experience is any guide, that is certainly true. If we present value all that, it might make the whole effort worthwhile. But Prof. Morris also adds that the real sector of the UK economy may have lost out in the pursuit of London‘s interest as a financial centre. If we subtract that, where does that leave us?

Posted at 12:05 on Thu, 28 Jul 2005     View/Post Comments (1)     permanent link

Mon, 11 Jul 2005

Overpricing of Emerging Market CDS?

A recent IMF working paper (Manmohan Singh and Jochen Andritzky (2005), “Overpricing in Emerging Market Credit-Default-Swap Contracts: Some Evidence from Recent Distress Cases”, IMF Working Paper 05/125) claims that there is significant overpricing of Credit-Default-Swaps on emerging market sovereigns.

The authors claim that the market prices CDS on an assumed recovery assumption of 20%. Under this assumption, the CDS spread can be used to compute an implied probability of default. The authors then show that even during periods of financial distress and restructuring (for example Argentina) the cash market price of the distressed bonds (even the cheapest to deliver bond) is well above 20% of par (it is typically 40% of par). The authors then compute an implied recovery rate from the cash market price of the cheapest to deliver bond by assuming the implied probability of default computed from the CDS spread under the 20% recovery assumption. They then compute the theoretical CDS spread using the implied probability of default from the CDS market and the implied recovery rate assumption from the cash market. This theoretical CDS spread is below the actual CDS spread.

It is difficult to understand what this whole exercise really proves. Yes, it does show that emerging market sovereign CDSs should not be priced under a 20% recovery assumption. Perhaps, it also shows that there was a divergence between the CDS market and the cash market — either the CDS was overpriced or the cash market was underpriced. To arbitrage this difference away, it would be necessary to short the cheapest to deliver bond in the cash market. This is where the first author’s earlier paper (Manmohan Singh (2003) “Are Credit Default Swap Spreads High in Emerging Markets? An Alternative Methodology for Proxying Recovery Value”, IMF Working Paper 03/242) throws some light. In that paper, Manmohan Singh explains that the cheapest to deliver bonds were squeezed and were on special repo. Moreover, the sovereign itself was trying to push up the price of the cheapest to deliver bond by buying up as much of it as possible. The price of the cheapest to deliver bond rose during the period of distress. All this points to a very different conclusion — that the cash bonds were overpriced. If so, it is not that the CDS spreads were too high but that the cash bond yields were too low.

Posted at 08:35 on Mon, 11 Jul 2005     View/Post Comments (0)     permanent link

Fri, 01 Jul 2005

Reputational Cost of Cooking the Books

Karpoff, Jonathan M., Lee, Dale Scott and Martin, Gerald, “The Cost to firms of Cooking the Books” (June 14, 2005) provide an interesting analysis of the penalties for financial misreporting in the United States.

They claim that the legal penalties are dwarfed by what they call the reputational penalty imposed by the market: “the reputational penalty is twelve times the sum of all penalties imposed through legal and regulatory processes.”

The difficulty is in the way that the reputational penalty is estimated. The authors take the drop in market value of the firm when an investigation is announced and subtract from this the monetary penalty imposed by the regulator as well as the amount paid in any class action suit. They then proceed to subtract the valuation impact of the accounting write-off required to reverse the financial misreporting. What is left is the authors’ estimate of the reputational penalty imposed by the market. The authors assume that the valuation impact of the accounting write-off is captured by multiplying the amount of the write-off by the price to book ratio.

consider the hypothetical example of an all-equity firm that has book value of assets equal to $100 and a market-to-book ratio of 1.5. The market value of the firm’s assets, and its shares, is $150. Assume the company then issues a misleading financial statement that overstates its asset values by $10. If the firm's market-to-book ratio stays the same, its share values will increase temporarily by ($10 x 1.5) to $165. But when the financial misrepresentation is discovered, the book value will be restated by $10, back to $100. If there is no other impact, the market value will fall by $15, back to $150. Thus, a $10 restatement in the firm’s books implies a $15 change in the market. This $15 drop in market value is what we seek to capture with the accounting write-off effect.

This estimate can be badly wrong because the accounting restatement can change the price to book ratio itself. A restatement that is relatively small in relation to the net worth of a company can make a large difference to the growth rate in earnings. If this changes the estimate of future growth opportunities, then the price to book ratio can fall dramatically. This is because the price to book reflects the relative importance of the present value of growth opportunities (PVGO) in relation to the present value of continuing operations (PVCO) and a relatively small change in growth rates can make a large change to the PVGO.

This is best seen in the context of a constant dividend growth model. Under this model, the market capitalization of the company is given by E(1-b)/(k-g) where E is the earnings of the company next year, b is the fraction of earnings that is retained (1 minus the payout ratio), k is the cost of equity and g is the growth rate in earnings and dividends. In this model E/k is the present value of continuing operations (PVCO) and is a rough estimate of what the book value would be. The balance is the present value of growth opportunities (PVGO).

Let us consider a numerical example. If last year’s earnings are $108 million, the dividend payout ratio is 40%, the cost of equity is 10% and the growth rate is 8%, the dividend growth model predicts a market capitalization of about $2.3 billion since the projected earnings next year are $116 million and the price earnings multiple (1-b)/(k-g)is 20. Of this, the PVCO is only about half ($108 million discounted at 10%), while PVGO accounts for slightly more than half. Assuming that book value is a rough approximation to the PVCO, the price to book for this company would be about 2. Consider an accounting restatement that changes the earnings last year from $108 million to $106 million. as against $100 million the year before last. This reduces the observed growth rate from 8% to 6%. If the market regards 6% as the true estimate of future growth, then the price earnings multiple would fall to 10 and the price to book ratio would fall to 1 as the PVGO simply vanishes. (The 6% growth amounts to only the required 10% return on the 60% of earnings that is retained and ploughed back into the business every year.) The market capitalization would then halve from $2.3 billion to $1.1 billion or a loss of over $ 1 billion. As against this, the authors’ estimate of the valuation effect would be $2 million times the price to book ratio of 2 or $4 million. The true valuation effect is then about 250 times what the authors estimate it to be.

If this is so and the bulk of the alleged reputational penalty is actually a valuation effect of the restatement itself, then the author’ findings can be interpreted very differently. The legal penalties are only a small fraction (only 1/12 or around 8%) of the true losses suffered by investors. Considering that creditors recover about 60% of the value of a defaulted bond, this is a very poor track record

Posted at 16:38 on Fri, 01 Jul 2005     View/Post Comments (0)     permanent link