Thursday, September 29, 2016

Concentration: the case for putting all your eggs in one basket

This article was first published in the Financial Times, FTfm, on September 30, 2013; Co-author: Khemchand Sakaldeepi

Is diversification the best way to invest in the market today? Not really. The portfolios of major investors worldwide make the case for another, often-ignored, strategy: concentration. Business schools need to refrain from pushing the merits of diversification without highlighting the efficacy of concentration.

“Do not put all your eggs in one basket. Diversify.” In 1952, investment aspirants received this clarion call from Harry Markowitz, a US economist and Nobel laureate. Peter Lynch, the famous US businessman and stock investor, “never saw a stock he didn’t like” and was a great proponent of portfolio diversification. While managing the Magellan fund, at the peak of his career, Mr Lynch’s portfolio had more than 1,000 stocks. To date, portfolio diversification remains the most important lesson taught to students of investment and risk management. The concept is a common thread in the investment approach of most fund managers and investors.

However, if we look at the portfolios of the rich and famous, they are, surprisingly, mostly concentrated. Several great investors, spread across geographies, have very concentrated portfolios. Warren Buffett, George Soros, Rakesh Jhunjhunwala and many others are renowned proponents of portfolio concentration. To Mr. Buffett, over-diversification presented a “low-hazard, low-return” situation and thus he dismissed it. A concentrated portfolio pivots on the absolute conviction of the investor in his or her stocks and his or her risk appetite.

A diversified portfolio, on the other hand, works well if the investor is optimistic about the stock, but wary of the associated risk. Investors like the first billion-dollar Indian investor, Mr. Jhunjhunwala, walk a fine line between the two.

John Maynard Keynes, the influential British economist, was another staunch supporter of concentration. “As time goes on, I get more and more convinced that the right method in investment is to put fairly large sums into enterprises which one thinks one knows something about and in the management of which one thoroughly believes,” he once said.

Mr. Buffett, echoing Benjamin Graham, the father of “value” investing, says he does not just buy an insignificant thing that bounces by a small percentage every day on the stock market. He buys part of a real business and thinks like the owner of a business would.

Mr. Buffett says: “Wide diversification is only required when investors do not understand what they are doing.” Bruce Berkowitz, founder of Fairholme Capital and a leading “value” proponent, adds that just a handful of significant positions are enough to do unbelievably well in a lifetime.

In 2012, the results of a study from the Paul Woolley Centre for the Study of Capital Market Dysfunctionality, University of Technology Sydney, showed that if skilled fund managers invested in concentrated portfolios, they would improve their performance markedly as compared with the portfolios that they would build under the compulsion to diversify. Despite mitigating stock-specific risks, the method of diversification cannot fortify the portfolio against market risks.

Advocates of concentration also opine that building or creating wealth with a diversified portfolio is difficult, unless the entire market is experiencing a bull phase and all the stocks in the portfolio are performing well. Even then, you may not get the full advantage of a multi-bagger as your investment in that particular stock would be just a fraction of your entire portfolio. The anti-diversification camp proposes that to generate wealth some concentration is required, provided people know how to assess their risk appetites and simultaneously pick winning stocks.

Fund managers today are caught in a catch-22 situation. Is wealth generated first by diversification and then maintained through concentration or vice versa? Knowing that concentration has been the mantra for success for most investment gurus, is it savvy to jump on the “diversification bandwagon” by adhering to popular belief? Awareness of such dilemmas and seeking clarity on them is essential for future managers.

It is, thus, time for business schools to introduce concentration as an important strategy in wealth creation, management and enhancement. Special attention needs to be given to this in business pedagogy, as the training of financial advisers and finance students will remain incomplete if it is restricted to the hallowed realm of diversification as the only plausible investment strategy.

Monday, September 12, 2016

Mobilizing Data and Analytics in Malaysia

The interview was first published by the Global Association for Risk Professionals on September 09th 2016

Insurers turn to predictive and anomaly-detection techniques as fraud losses cut into profitability and cause regulatory concerns.

Mounting motor insurance claims have been a source of worry for general insurers in Malaysia. They suffered a net loss ratio for Motor Act business of 219.6% in 2015, and an above-100% combined ratio for the total motor portfolio. The losses are attributable to high frequency of third-party bodily injuries, rising accident rates and medical costs, as well as fraudulent claims (Source: ISM Insurance Services Malaysia’s Statistical Yearbook.) Industrywide efforts are being made to control the fraudulent claims payout. A fraud detection and prevention system of Insurance Services Malaysia (ISM) is part of the industrywide effort to control payouts of fraudulent claims.

ISM was conceptualized in 1998 by the General Insurance Association of Malaysia (PIAM), initially to establish the Malaysian Insurance Rating Organization. In 2003, MIRO and the MIS department were merged to form the ISM department, and the scope of the project was expanded to include anti-fraud and IT-related services. ISM Berhad was incorporated in 2005 and today provides an infrastructure of databases and analytics that allows members to make informed decisions and support a liberalized pricing environment, build competencies in quantitative underwriting and technical pricing, and increase efficiencies in operations. The shared information and capabilities are accessible online to enhance fraud detection.

Mahendran (Mahen) Samiappan, the chief executive officer of ISM, discusses in this interview with Dr. Nupur Pavan Bang the general insurance industry in Malaysia and efforts by the organization to curtail fraud.

What is the state of the general insurance market in Southeast Asia, and particularly Malaysia?
Motor Insurance is the largest portfolio in the region, followed by property. Health insurance is generally sold as a rider with life insurance policies. So in comparison to the life segment, health is not a very big business for the general insurance market. Health is also sold as unit linked policies by the life insurers.

In terms of technology and distribution, Singapore would be the leader, followed by maybe Malaysia and Thailand. Everyone acknowledges that Malaysia has good regulations due to the active role played by Bank Negara Malaysia as the regulator of this industry. We have averaged a growth rate of 6% to 8% in general insurance consistently. In 2015 the industry experienced much lower growth, 2.7%, and 2016 is expected to have a low growth rate as well.

What are some of the major challenges in Malaysian insurance?
The biggest challenge is pertaining to the motor portfolio. Motor comprises 47% of the total portfolio. It is still under the tariff regime, for both the third party (act or mandatory) as well as the own damage (non-act) cover. The act business endures heavy losses as a result of rising claims-cost bleeding. The current tariff was set out in 1978 and has not been revised since then.

Bank Negara Malaysia has put in place a road map for gradual liberalization of motor insurance tariffs. By July 2017, non-act business will be de-tariffed, and gradually the act business may also see certain adjustments in prices. The flexibility to price is important for the Insurers, but it may lead to price wars, as has been witnessed in some other countries. That may still mean that the portfolio remains loss making.

Is the market ready to price?
The larger players, who have backing from their foreign partners with considerable underwriting and technical know-how, are ready for risk-based pricing. The largest insurer would have about 15% of the market share and, with that kind of data, coupled with their capabilities, can do the pricing.

The smaller companies may not have adequate data, and this is where ISM comes in. ISM was established to support the industry in the de-tariffed environment. So we will definitely play our role and carry out our responsibility to support the industry in this transition. The regulator has also been aggressively going around and assessing the readiness of the insurers.

Does Malaysia face the uninsured-vehicle challenge that some markets do? Uninsured vehicles not only result in loss of premiums for the insurers, they also have huge economic impact on the uninsured vehicle owners and victims of accidents.
Not really. The Road Transport Department and insurance databases are linked. So if a vehicle does not have the mandatory third-party policy, road tax will not be issued by the department. Road tax in Malaysia is annual, and almost all vehicles would have at least the act policy.

For the victims of road accidents, ISM provides a Vehicle Information Exchange service – if the vehicle registration number is captured, the details of the insurer of the vehicle can be obtained.

How is the industry tackling its leakage problems?
It is true that leakages from premiums as well as claims are plaguing the insurers. It is plain fraud. In its road map, the regulator has clearly stated that they want to control and manage fraud. They don’t want the insurers to be lax on claims and then keep adjusting (increasing) the premiums to cover the losses. The focus is on fraud and data quality.

ISM’s Central No Claims Discount (NCD) database is already being used by the insurers to plug leakages at the application stage. Over the years, the NCD database has evolved into a system called Claims & Underwriting Exchange (CUE), which provides alerts to the insurers based on certain business rules. For example when an old vehicle moves from act only policy to a comprehensive policy, it is unusual, and hence an alert is sent to the insurer. These initiatives have helped to some extent.

Can you elaborate on fraud at the application stage?
Fraud at the application stage can be perpetrated by a customer providing a bogus identity or falsifying records like the driving license, use of the vehicle, incorrect claims history, etc. There is a strong linkage between claims fraud and fraud committed at the application stage. Reducing application fraud can significantly decrease the exposure to certain types of claims fraud. Stopping fraud at the application stage saves investigation, claim adjudication, litigation and recovery expenses.

How is the industry planning to deal with such fraud?
ISM is working on a platform with an objective for comprehensive fraud detection and prevention. Using analytics techniques such as predictive modeling, link analysis, anomaly detection and text mining, claims will be scored, and certain claims with high potential of fraud will be highlighted based on business rules and analytics. The insurers can allocate resources to investigate the highlighted claims and thereby make more effective use of available resources.

However, analytics and models are only as good as the data. So a right mix of analytics with prudence, diligence and judgment should be applied by the insurers while processing applications and claims.