THE ECONOMICS OF FINANCIAL MARKETS
Do
changes in UK interest rates affect the London stock market? Or are the
movements dominated by what is happening on the New York stock exchange? This
study quantifies such effects by examining the extent to which movements in
London stock market prices depend on economic and financial information.
Methods
traditionally used to model the movements in economic or financial variables are
linear techniques, which assume that individual factors always have the same
effects. Therefore, such models imply that increases and decreases in a relevant
factor have symmetric effects on stock market prices.
These
researchers relax this assumption by using a flexible nonlinear technique,
namely the smooth transition regression model. A range of macroeconomic and
financial variables are examined for their possible effects on London stock
market price movements in both linear and nonlinear models, with the New York
index emerging as the key variable for explaining the London FT index.
The
data used in this study are quarterly, since it is such medium-term movements
that may best reflect the impact of underlying economic and financial factors on
the stock market. The period examined is from 1967 to 1996, and hence includes a
number of periods of financial turbulence, including the mid-1970s and the stock
market crash of 1987.
ENDS
For
Further Information:
contact Professor Denise Osborn on 0161-275-4861 (email: denise.osborn@man.ac.uk);
or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095
(email: romesh@compuserve.com).
Recent government pensions policy, such as the introduction of stakeholder pensions, has encouraged people to save through defined contribution schemes invested in tracker funds on the basis that there is little evidence that active fund management delivers superior performance. New research by Professor Ian Tonks, to be presented at the Royal Economic Society’s Annual Conference this week, challenges the view that fund managers cannot outperform the market.
Tonks
examines the performance of fund managers responsible for managing the equity
investment decisions of a large sample of UK pension funds. He finds strong
evidence of persistence in ‘abnormal’ performance at the one-year time
interval. Using a large sample of 1,750 UK pension funds over the period 1983-97
provided by CAPS, he calculates measures of excess returns to equity returns on
pension funds, and then examines whether these excess returns are attributable
to individual fund management houses.
The
research also examines whether past performance of pension fund managers is
linked to their future performance. This is done by ranking fund managers’
past performance over alternative time periods (the past three months, the past
12 months and the past 36 months), and then evaluating whether past performance
is related to performance over the subsequent evaluation periods (the next 3
months, the next 12 months and the next 36 months).
Tonks
finds evidence of significant persistence in the performance of fund managers at
the 12-month time horizon using a number of different consistency tests, as well
as some evidence of persistence at other time intervals. The returns on a zero
investment portfolio of a long position in a portfolio of fund managers that
performed well over the previous 12 months and a short position in a portfolio
of fund managers that performed poorly, would have yielded an annualised
abnormal return of 1.56%. This abnormal performance is significantly higher than
the annual fund management costs reported in the Myners Report (2001), which are
typically around 40 basis points (0.4%)
Previous
work on UK pension funds has found only weak evidence of fund manager
persistence. Tonks suggests that this might be due to ‘survivorship bias’ in
these samples, which disguise true persistence. He argues that these earlier
studies may have induced a selection bias by restricting the data sample to the
same fund manager over a long time period, and this survivor bias may have
reduced the level of persistence in the sample. In fact using his dataset, with
the restriction that only long-lived funds with the same fund manager are
included, reduced the return on a zero investment portfolio by half.
One
caveat is in order. An implication of these results might be that pension fund
manager mandates should be set up on a yearly basis. But this would ignore the
substantial transactions costs involved in shifting a pension funds’ assets
from one fund manager to another, at such regular intervals. Tonks’ analysis
has not factored in the costs of shifting pension fund managers on a regular
basis.
ENDS
Notes
for Editors:
‘Performance Persistence of Pension Fund Managers’ by Ian Tonks will be
presented at the Royal Economic Society’s 2002 Annual Conference at the
University of Warwick on Wednesday 27 March.
Tonks
is Professor of Finance at the Centre for Market and Public Organisation,
Department of Economics, University of Bristol, 8 Woodland Road, Bristol BS8
1TN.
For
Further Information: contact
Ian Tonks on 0117-928-8435 (home: 01225-422196; fax: 0117-928-8577; email: I.Tonks@bristol.ac.uk);
or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095
(email: romesh@compuserve.com).
Stocks
of UK companies that issued profits warnings deliver returns that are on average
22% more than the FTSE All Share index in the year beginning 12 months after the
warning. That is the key finding of new research by Professor George
Bulkley, Richard Harris and Renata Herrerias, which they presented at the Royal Economic
Society’s Annual Conference on Tuesday 26 March. Their analysis of market
reactions to profit warnings and subsequent annual earnings announcements
suggest that investors under-react to the former but over-react to the latter.
Many
investors believe that stock markets do not respond rationally to new
information. Some believe markets over-react to news, and hence attempt to beat
the market by using ‘contrarian’ trading strategies - for example, buying
stocks that have reported bad news. Others believe markets are too slow to
revise their existing views on a company so that markets under-react to news.
These investors buy stocks that have just reported good news on the assumption
that they still have further to rise, sometimes called a momentum strategy.
This
study tests which of these views is correct using data on returns following
profit warnings. The authors examine returns over the two years following a
profit warning for all UK companies that issued profit warnings between
September 1997 and December 1999. They find that:
·
Returns on the day of a warning are on average minus 16%. The question is
whether this is an over- or under-reaction.
·
Over the next six months, these companies on average lose a further 5% of
their value, supporting the under-reaction view and the momentum trading
strategy.
·
But tracing returns for a full two years after the warning, a different
picture emerges. Stocks that issued warnings deliver on average 22% more than
the FTSE All Share index in the year beginning 12 months after the warning.
This
suggests that markets under-react to a profit warning in the sense that they
underestimate its implications for the forthcoming actual earnings announcement.
Since profit warnings precede earnings announcements by a matter of months,
negative returns follow over a six month horizon as the market comes to see that
it did not sufficiently revise down its earnings forecast after the warning.
But
what explains the striking returns after 12 months? Psychologists report that
individuals are too ready to identify trends when shown data that is simply
random, with no underlying trend. They forecast the future by extrapolating the
trend they think they have identified, which results in systematically mistaken
forecasts. The same factor seems to be at work in stock markets. The market
gives too much weight to a single year’s earnings announcement. It downgrades
its forecasts of future profits too much in response to one bad year.
There
are economic mechanisms at work to restore earnings after an unusually bad year.
Companies that have reported bad earnings may set in train cost saving measures
that might have been unacceptable to the management, and the work force, in
normal times - for example, Marks and Spencer in 2001. In other cases, senior
management may be replaced - for example, Marconi.
On
average, the market underestimates these responses, so that profits in the year
following the warning year are better than expected. We might expect that
approximately 12 months after the profit warning, the market starts to get
signals about the following year’s earnings, thus explaining these results.
In
summary, the market over-reacts to the annual earnings announcement. Given its
(incorrect) beliefs about the significance of annual earnings results for
long-run profits, it is too slow to revise its expectations about the current
year’s earnings following a profit warning.
ENDS
Notes
for Editors:
‘Stock Returns Following Profit Warnings: A Test of Models of Behavioural
Finance’ by Professor George Bulkley, Richard D.F. Harris and Renata Herrerias
was presented at the Royal Economic Society’s 2002 Annual Conference at the
University of Warwick.
The
authors are at the University of Exeter.
For
Further Information:
contact George Bulkley on 01392- 263214 (email: I.G.Bulkley@exeter.ac.uk); RES
Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email:
romesh@compuserve.com).
HOW
USEFUL ARE SECURITY ANALYSTS?
Security
analysts play an important role in reducing ‘agency costs’ in UK firms –
the costs arising from the fact that senior managers naturally tend to put their
own interests ahead of the goal of maximising value for their shareholders. But
they are considerably less effective for larger and more diversified firms than
for smaller and more focused firms. What’s more, security analysis seems to
have a greater influence on the value of smaller firms than that of larger
firms.
These
are the central conclusions of new research by John
Doukas, Phillip McKnight and Christos
Pantzalis, presented at the Royal Economic Society’s Annual Conference on
Monday 25 March. Their findings suggest that the usefulness of security analysis
tends to diminish with the size and diversification of the firm despite the fact
that more resources are spent on the acquisition and evaluation of private
information for larger firms.
Security
analysts play an important role for the financial markets in their collection,
processing and dissemination of information to interested users. They normally
conduct firm- and industry-specific research, the findings of which are conveyed
to the public in the form of annual reports and earnings per share announcements
that are predictive in nature. Brokerage firms employ thousands of
security analysts and make enormous investments in analysing such information in
their stock recommendations.
Stock prices
tend to respond more to changes in analysts' forecasts of earnings than they do
to changes in earnings themselves, suggesting the usefulness of analysts’
earnings forecasts. One economic
benefit of security analysis is that it increases the transparency of the firm
by extracting and disseminating firm-specific information to existing and
potential investors. In that capacity, security analysis assists investors in
their decision-making processes.
Jensen
and Meckling’s (1976) ‘agency theory’ is perhaps one of the more important
contributions to the modern finance literature. Its key idea is that security
analysis should reduce the agency costs associated with the separation of
ownership and control. Job perks, shirking, and decisions taken to maximise
managers’ utility are just a few forms of agency costs and these costs can be
boundless unless managerial actions are properly monitored.
Until
now, the role of security analysts in reducing agency costs has not been tested
in a UK environment, which is surprising given the growing reliance on them to
provide investment guidance. Given the perceived importance of security analysts
coupled with the dilemma surrounding managerial monitoring, these researchers
decided to examine just how important their influence may be on both the value
of the firm and any agency cost born by the firm. The study used data from
Multex-Global Estimates, the second largest global provider of analyst earnings
forecast estimates.
The
evidence they produce is consistent with the view that security analysis has
monitoring capabilities in reducing agency costs. But security analysis is
considerably less effective in restricting managers’ non-value maximising
behaviour for larger and more diversified firms than for smaller and more
focused firms. Moreover, while firm value is a positive function of security
analysis, security analysis also seems to exert greater influence on the value
of smaller rather than larger firms. This suggests that the usefulness of
security analysis tends to diminish with the size and diversification of the
firm despite the fact that more resources are spent on the acquisition and
evaluation of private information for larger firms rather than for smaller
firms.
Overall,
in accord with the US evidence on the monitoring role of security analysis,
these findings suggest that UK information intermediaries, such as security
analysts, are to a large extent effective in reducing agency costs associated
with the separation of ownership and control. The results are also consistent
with those found by Doukas et al in a recent US study suggesting that the
monitoring role of security analysts is not limited to the US capital market
environment.
ENDS
Notes
for Editors:
‘Security Analysis, Agency Costs, and UK Firm Characteristics’ by John A.
Doukas, Phillip J. McKnight, and Christos Pantzalis was presented at the Royal
Economic Society’s 2002 Annual Conference at the University of Warwick.
Doukas
is in the Department of Finance, Stern School of Business, NYU and the School of
Business and Public Administration, Old Dominion University, Norfolk, VA
23529-0218. McKnight is Distinguished Professor of Finance, Department of
Accounting and Finance, Cardiff Business School, Cardiff CF10 3EU. Pantzalis is
in the Department of Finance, College of Business Administration, University of
South Florida, Tampa, FL 33620-5500.
The
research was financially support by the Leverhulme Trust. Most of the data was
provided by Multex-Global Estimates.
For
Further Information:
contact Phillip McKnight on 02920-876804 (fax: 02920-876804; email mcknightpj@cardiff.ac.uk) ;
John Doukas on +1-757-683-5521 (fax: +1-757- 683-5639; email: jdoukas@stern.nyu.edu; Christos Pantzalis
on +1-813- 974-6326 (fax: +1-813- 974-3030; email: cpantzal@coba.usf.edu); or RES Media
Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).
HOW
RELIABLE ARE GLOBAL CREDIT RATING AGENCIES?
How
reliable are the pronouncements of global credit rating agencies like Moody’s,
Standard & Poor’s and Fitch-IBCA? New research by Giovanni
Ferri
and Li-Gang
Liu,
presented at the Royal Economic Society’s Annual Conference on Wednesday 27
March, finds evidence on the information content of firm ratings that will not
reassure investors and regulators. They conclude that rating agencies must
improve their rating quality especially in emerging economies in order to
measure up to their increased global presence and influence.
Investors,
regulators and other market participants are increasingly relying on the ratings
issued by global credit rating agencies to assess issuers’ quality. The
reliability of these ratings has come increasingly under scrutiny. Rating
agencies have been criticised for failing to forewarn markets when things were
going awry for issuers. For example, why did major rating agencies downgrade
Enron to junk status only days before the company took its books to the court?
And why were they so late to downgrade Indonesia and Korea during the 1997
crisis?
In
other words, does the information content of the ratings measure up to
issuers’ risks? This is a fundamental concern especially in light of growing
scepticism on the truthfulness of companies’ accounting statements. And if the
quality of ratings needs improvement even in the most developed markets, how
desperate is the situation for companies in emerging economies? The study
addresses three key questions:
Is
the information content of company ratings on the same par in emerging and
developed economies?
The
researchers calculate to what extent individual firm ratings reflect the
performance-based rating that can be assigned to firms from the economic and
financial indicators reportedly used by rating agencies. They explore whether,
in fact, ratings simply mimic the creditworthiness of the country in which they
are located - as given by the sovereign rating - thus incorporating no
additional information for investors.
The
results indicate that firm ratings depend closely on respective sovereign
ratings in emerging economies though not in developed countries. Projecting the
ratings on a numerical ladder from 100 to 0 - Aaa=100, Aa1=95, Aa2=90 … down
to Caa=5, C=0 - the research finds that:
·
When the sovereign rating is downgraded by five points, the average
downgrade for firms is about four points in emerging economies while there is no
downgrade in developed countries.
·
In contrast, if the firm’s performance-based rating improves by five
points, its actual rating increases by six points if it is a developed country
firm and only by half of that if it is an emerging economy firm.
Is
the information content of firm ratings lowered in emerging economies by the
‘country ceiling effect’, whereby private ratings are bound not to exceed
their sovereign rating?
The
robustness of this result is then checked to see whether it was just driven by
the ‘country ceiling effect’. On the contrary, in emerging economies, firm
performance indicators are found to be irrelevant even for firms whose rating
would lie anyhow way below their sovereign’s.
To
what extent do the failings of firm ratings reflect poorer information quality
standards?
The
study examines whether the lower information content of firm ratings in emerging
economies results from the fact that information quality standards are also
generally worse there than in developed countries. In other words, if emerging
economy firms are more opaque, then it is proper for rating agencies to rely
less on their performance indicators.
The
linkage between firm level rating and performance is indeed weaker in countries
where company data are less transparent. But even taking this into account,
cross-country indicators of information quality cannot fully explain why the
information content of company ratings is so disproportionately lower in
emerging economies.
These
findings have disturbing implications for the new Basel Accord among G-10
countries, correlating banks’ capital adequacy requirements to external credit
ratings. The question is even murkier for emerging economies. Thus, rating
agencies must improve their rating quality especially in emerging market
economies in order to measure up to their increased global presence and
influence.
ENDS
Notes
for Editors:
‘Do Global Credit Rating Agencies Think Globally? The Information Content of
Firm Ratings around the World’ by Giovanni Ferri and Li-Gang Liu was presented
at the Royal Economic Society’s 2002 Annual Conference at the University of
Warwick. Ferri is in the Department of Economics, University of Bari, Via C.
Rosalba, 53, 70124 Bari, Italy; Liu is at the Asian Development Bank Institute,
Tokyo, Japan, respectively. The views are those of the authors alone and should
not be attributed to ADB, its board of directors, or the countries they
represent.
For
Further Information:
contact RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095
(email: romesh@compuserve.com);
or Giovanni Ferri via fax: +39-080-5049149 or email: g.ferri@dse.uniba.it.
KEY
DRIVERS OF INDIVIDUAL OWNERSHIP OF
STOCKS
AND MUTUAL FUNDS
Ownership
of risky assets like stocks and mutual fund by private households has been
rising significantly in many Western countries since the early 1990s. New
research by Rob Alessie, Stefan Hochgürtel
and
Arthur van Soest, presented at the
Royal Economic Society’s Annual Conference on Tuesday 26 March, explores the
key influences. Their data reveal that:
·
The probability of owning stocks increases significantly with income
while the probability of owning mutual funds does not. Stocks carry fewer
transaction costs than mutual funds for large investors since they are less than
proportional to the amounts held. Mutual funds provide instant diversification
at the cost of a premium, and are particularly attractive for small investors
with little financial knowledge.
·
The self-employed have a 25 percentage points higher probability of
holding stocks than employees, while they do not have a different ownership rate
of mutual funds. An explanation could be that the self-employed are interested
in specific stocks to hedge against their larger income uncertainty. It is
unlikely that the behaviour of the self-employed is due to preferences for less
diversification.
·
In general, owning one asset type this year is a strong predictor for
owning the same asset next year. Adjustment costs of buying or selling the asset
– either real or perceived - can explain this.
·
Relatively often, people either own both of these types of assets or
neither of them. The positive correlation between ownership of one asset and
(lagged) ownership of the other asset is entirely explained from variation in
household characteristics, tastes and technology. Factors behind this may be
joint elements in monitoring or other holding costs, or benefits of
diversification that cannot be attained by mutual funds alone (since these
typically invest in certain sub-samples of stocks).
· There is no evidence that households substitute one type of asset for the other, or that ownership of one type leads to more financial knowledge and a larger probability of buying the other asset type. In contrast, there is some evidence of a negative effect of owning one type of asset on buying or keeping the other type. Adjustment costs, which imply that those who have acquired one specific asset will tend not to reallocate their money to the other type of asset, can explain this. Such adjustment costs will comprise monetary transaction costs of portfolio adjustment, and possibly also non-monetary or perceived costs components, reflecting the required effort, the costs of acquiring information, etc.
An
important global trend in households' finances over the past decade has been
rising ownership rates of risky assets. The percentage of private households
owning risky financial assets has increased substantially during the 1990s in
many countries. Between 1989 and 1998, this fraction rose from 31.9% to 49.2% in
the United States, and from 12.0% to 22.1% in Italy. Particularly noteworthy is
the strong growth in equity-based mutual funds. Such trends are also observed in
the Netherlands, one of the few countries for which household level data are
available that make it possible to study individual dynamic portfolio behaviour.
Because
of its potentially wide-ranging implications for the allocation of risk in
financial markets and for the macro economy, policy-makers need to understand
the behavioural factors that drive ownership dynamics at the individual
household level. To do so, these researchers analyse a representative sample of
Dutch households that are followed over time.
The research analyses the interactions between the demands for individual stocks and mutual funds, the two most important risky asset types. The methodology makes it possible to distinguish between various reasons that may explain why people relatively often either own both of these types of assets or none of them. It explains ownership from observable characteristics such as age, labour market position, family composition, etc., and previous ownership. The conclusion is that adjustment costs - either real or perceived - are a likely candidate for explaining the observed persistence of ownership of both types of assets.
ENDS
Notes
for Editors: ‘Ownership of
Stocks and Mutual Funds: A Panel Data Analysis’ by Rob Alessie, Stefan Hochgürtel
and Arthur van Soest was presented at the Royal Economic Society’s 2002 Annual
Conference at the University of Warwick.
The
study was presented by Hochgürtel, who is Finance and Consumption Chair,
European University Institute, Florence, Italy.
For
Further Information: contact
Stefan Hochgürtel on +39-055-4685-302 (fax: -202; email: stefan.hochguertel@iue.it);
or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095
(email: romesh@compuserve.com).
MERGERS
AND ACQUISITIONS IN THE BANKING INDUSTRY:
LESSONS
FROM ARGENTINA
How
should policy-makers react to consolidation in the banking industry? New
research by María Eugenia Delfino,
presented at the Royal Economic Society’s Annual Conference on Wednesday 27
March, suggests that policies limiting mergers and acquisitions among banks may
be misdirected if consolidation and the resulting concentration are motivated by
cost economies. Her study of the Argentine banking industry indicates that banks
charge higher than competitive prices for their services, but that such costs
have been more than offset by efficiency gains, benefiting customers as well as
the banks themselves.
Argentina
presents an interesting case for studying the consolidation of the banking
industry that is occurring all around the globe. Several reforms implemented
during the 1990s led to changes in the structure of this industry towards
greater consolidation. As elsewhere, this process raises important policy
concerns:
·
On the one hand, it can be argued that banks merge to exploit cost
economies, which could benefit consumers if these efficiency gains translate
into lower interest rates on loans and higher rates on deposits.
·
On the other hand, it can be argued that banks gain market power from
merging, which could allow them to adopt pricing policies less favourable to
consumers.
The
findings of this study indicate that banks in Argentina charge higher than
competitive prices for their services. But the study also indicates that such
costs have been more than offset by cost efficiency gains, which resulted in
economic benefits not only for banks but also for consumers. These findings
suggest that policies forcing downsizing in highly concentrated industries may
be misdirected if consolidation is motivated by cost economies.
These
findings are of considerable interest for policy-makers:
·
First, they indicate that banks in Argentina gained market power during
the 1990s, possibly due to the consolidation process.
·
Second, they suggest that small-sized banks have exerted a higher degree
of market power pricing than larger institutions, which could be explained by a
different clientele.
·
Third, they further show that banks obtained efficiency gains derived
from both a larger scale of operations and associated technological advances.
·
Fourth, they also suggest that the consolidation process may proceed
since most banks could still gain from an increase in their size.
·
Finally, they show that despite market power increased during the decade,
consumers benefited because the increase in the size of banks together with the
technological advances led to lower costs and more favourable prices.
This
evidence implies that policies limiting mergers and acquisitions among banks may
be misdirected if consolidation and resulting concentration are motivated by
cost economies.
Mergers
and acquisitions among financial institutions are occurring at a fast pace in
the United States; they may increase in the near future in Europe under monetary
union; and they may be part of the solution to problems of financial distress in
developing countries. Existing studies analyse the consolidation process in the
banking industry focusing either on the analysis of market power or on the cost
structure, rather than on a careful representation of both ‘blades of the
scissors’.
The
case of Argentina is interesting because several reforms implemented during the
1990s prompted the banking system to grow in a more competitive environment,
which in turn led to changes in the structure of the sector towards greater
consolidation. As a result of this process, which accelerated after the Mexican
crisis of 1994, the number of banks halved while concentration of deposits and
loans among the largest institutions increased sharply. This study is unique
because it analyses simultaneously the market and cost structure of the banking
industry using data for all retail banks operating in Argentina during most of
the decade.
ENDS
Notes
for Editors:
‘Consolidation and Competition: The Case of the Argentine Banking Industry’
by María Eugenia Delfino was presented at the Royal Economic Society’s 2002
Annual Conference at the University of Warwick.
Delfino
is a Doctoral Researcher in the Department of Economics, University of Warwick,
Coventry CV4 7AL.
For
Further Information: contact
María Eugenia Delfino on 024-7652-8415 (fax: 024-7652-3032; email: M.E.Delfino@warwick.ac.uk);
or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095
(email: romesh@compuserve.com).
Last updated
12th April 2002