SMSFs beat professionals… and other manager research

Ron Bird
Share on facebook
Share on twitter
Share on linkedin
Share on email

(Pictured: Ron Bird)

Professional investors may mock the stock selection of SMSF trustees for lack of diversity and heavy weighting to the banks, but a new study shows that such a strategy may well perform better than that of fund managers, both domestic and international.

The study, by Peter Swan and others at the University of NSW, claims to be the first to analyse counterparty trades for an entire market, in this case the Finnish sharemarket between 1995 and 2012. A key component in the outperformance of individual investors versus fund managers was their buy-and-hold tendency. Another key component was that individuals did not exhibit as much trend-following behaviour as the fund managers, especially international managers.

Swan was one of the presenters last week of new research at a workshop organised by Professor Ron Bird of the Paul Woolley Centre at the University of Technology Sydney (UTS), who also oversees PhD students in finance at the University of Waikato, New Zealand.

Swan said: “It’s outrageous to claim that individual investors trade too much. In Finland, at least, 40 per cent of all households do not trade their shares in any one year. The average turnover for households was 40 per cent a year. This compares with 1200 per cent for the foreign nominees [managers].”

He said the Finnish individuals typically held only one or two stocks, but although they had “no diversification”, they were better informed than both domestic and fund managers and were better able to market time.

“Householders are the winners against all opponents, followed by domestic institutions, with foreign investors always the losers. This explains the home-bias informational leakage,” he said.

“Foreign investors made a gift of between 20-25 billion euro during this period, which included a slump in Finland’s then-biggest stock, Nokia, and its subsequent takeover by Microsoft. Foreign managers dumped Nokia after the sell-off so that the takeover premium was mainly gained by local investors.

Questioned about the impact of this one situation on his overall results, Swan said that it did not alter the rankings of the three investor classes although it did reduce the gaps between them. He added that he believed studies in other countries would show similar results.

Finland has a unique dataset, however, with all shareholders having to register through a national registry, obtaining an ID for life, and the daily transactions recorded and available through Euroclear. About one million investors were recorded for the study, which covered 33 stocks. Foreign investors accounted for about 95 per cent of all trading in the Finnish market, mostly in Nokia.

Ron Bird and Danny Yeung, also of UTS and formerly the Paul Woolley Centre, made separate presentations on studies relating to fund manager abilities. Yeung examined persistence of performance and new ways to predict future outperformers, while Bird looked at stock selection performance through manager ‘hit rates’.

Yeung’s study identified managers according to factors other than past performance, including: active bets – their divergence from assigned benchmarks; style tilts; industry concentrations in their portfolios; and their returns gap – the actual returns at the end of a quarter versus the returns of the stocks they held at the start of the quarter.

He found it was possible to identify managers who outperformed by 2-3 per cent for the subsequent quarter. Yeung and his collaborators were now looking to make their model more “implementable” by looking further out than one quarter.

In Bird’s study, which used the same dataset and period, from 1991-2012, he compared the managers’ ‘hit rates’ versus that of their benchmark. The hit rate was defined as the proportion of stocks which outperform the benchmark. The benchmarks actually have a hit rate below 50 per cent for all time periods of one, three, six and 12 months, but the managers’ hit rates were even lower. Only 42.45 per cent of managers achieved hit rates better than their benchmarks.

Small-cap and value managers tended to have the best hit rates, Bird said. The hit rates were also closely related to fund performance, although portfolio construction (stock weightings) was not considered in the study.

Share on facebook
Share on twitter
Share on linkedin
Share on email