65 K ings way London W C2B 6TD Tel:+44(0)20 7831 0898 Fax:+44(0)20 7831 9975

w w w . i n v e s t m e n t u k . o r g

Investment Management Association is a company limited by guarantee registered in England and Wales.

Registered number 4343737. Registered office as above.

23 November 2011 kayreview@bis.gsi.gov.uk Professor John Kay Department for Business, Innovation and Skills Spur 2, Floor 3, 1 Victoria Street London SW1H 0ET

Dear Professor Kay

THE KAY REVIEW OF UK EQUITY MARKETS AND LONG-TERM DECISION MAKING

IMA represents the asset management industry operating in the UK. Our members include independent fund managers, the investment arms of retail banks, life insurers and investment banks, and the managers of occupational pension schemes. They are responsible for the management of £4 trillion of assets, which are invested on behalf of clients globally. These include authorised investment funds, institutional funds (e.g. pensions and life funds), private client accounts and a wide range of pooled investment vehicles. In particular, the Annual IMA Asset Management Survey showed that in 2010 IMA members managed holdings amounting to 40 per cent of the domestic equity market. In managing assets for both retail and institutional investors, IMA members are major investors in companies whose securities are traded on regulated markets. They value an efficient equity market that operates to benefit both companies and markets, and which enhances the long-term competitiveness of UK corporates and returns for savers. It is vital that the UK continues to attract international investment and maintains an internationally competitive fund management industry. We thank you for the opportunity to comment on this call for evidence and for attending the recent roundtable with a cross-section of managers from our membership. We set out in Annex 1 our detailed evidence under each of the questions and below our key points. Investors are long-term. IMA’s response to the Secretary of State’s call for evidence

last year noted that the great majority of investors in UK equities manage clients’ money in ways that seek to build value over the long-term1. Only by delivering sound performance over the long-term will the buy-side, asset managers, retain their clients. In recent years there has been a growth of alternative strategies that depend on short-

1 http://www.investmentfunds.org.uk/assets/files/consultations/2011/20110113-bis.pdf

2

term market trends and turn over portfolios more rapidly - markets need diversity to ensure liquidity and efficiency. However, the way long-term investors manage their assets has not changed - our detailed evidence is set out in question 1d.

Nor are asset managers incentivised to hold equities short-term in that they are remunerated on a percentage of the assets managed (a fixed percentage and sometimes a performance fee). Thus their interests are aligned with those of their clients and investee companies – the better companies perform, the better returns for clients and the better managers are remunerated. If a manager over-trades it has an

adverse impact on performance and revenues because of costs involved – see our evidence in question 6c.

Short-term company incentives. As regards investment decisions taken by companies, the incentives applying to Boards and, in particular, senior executives can be an important driver of company behaviour. Remuneration schemes are frequently based on short-term earnings and share price. We understand that even long-term incentive plans rarely extend beyond three years. The short tenure of certain executives - the average Chief Executive Officer tends not to serve for many years – may also not be conducive to fostering long term perspectives. Some have questioned whether the metrics most commonly used to assess performance, such as earnings per share or return on equity, are in fact aligned with the creation of long-term value. We think this is an area to which the Review could usefully devote attention – see 3c.

The costs and incentives of intermediation. The Review should, in our view, consider the potential for value to leak to intermediaries in the equity markets, which may be incentivised to encourage behaviour which is not necessarily in the long-term interests of companies. Corporate financiers and other advisers tend to be remunerated through fixed fees which are contingent on merger and acquisition deals being completed. They thus have an incentive to encourage the completion of such deals. But the evidence that this activity adds long term value is at best mixed. A further example is that investment bank advisers receive significant fees when companies raise capital, by underwriting and/or acting as agent. The Institutional Investor Committee, of which IMA is part, issued a report in 2010, the Rights Issue Fees Inquiry, and made a number of recommendations. The Review might consider whether more fees should be deferred with their payment dependent on the transaction securing the long-term success of the company as opposed to simply its completion – see 6d.

Integrity of the UK equity market. The internationalisation of the UK equity markets has increased the available capital and reduced its cost. However, there are concerns that the recent influx of certain overseas mining companies where the minimum free-float requirements were waived prior to listing and not reinstated may, if such practices persist, damage the integrity of the UK market. It is vital that the quality of a UK listing is maintained particularly as this is very often a passport to being included in the main FTSE UK Indices and accessing the funds of the large passive asset managers – see 9a.

Terms of reference. The terms of reference for the Review focus solely on the operation of the equity markets. Companies also access capital through issuing fixed income instruments whose holders can also impact a company’s long-term competiveness. For completeness, we consider the Review should have addressed other types of capital available to companies.

3

Please contact me if you would like clarification on any of the points in this letter or the attached, or if you would like to discuss any issues further. Yours sincerely

Liz Murrall Director, Corporate Governance and Reporting

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

4

IMA’s evidence in relation to the questions raised is set out below.

1. Whether the timescales considered by boards and senior management in

evaluating corporate risks and opportunities, and by institutional shareholders and asset managers in making investment and governance decisions, match the time horizons of the underlying beneficiaries.

Evidence is invited on: a. the relationship between reporting timescales and those used for internal

planning and appraisal; b. what timescales are used by companies in investment appraisal; and c. how companies review investment in intangible assets (e.g. corporate

reputation, workforce skills).

IMA considers that points a. to c. are for companies to answer in that they relate to their internal processes.

d. what timescales are used by equity investors, and in particular

institutional investors such as pension scheme trustees, who appoint fund managers in determining investment strategy.

The majority of equity investors be they investors in collectives, pension scheme trustees, or insurers, invest in equities by appointing an asset manager to invest on their behalf. Asset managers have two distinct strategies: active and passive management although the majority operate a mix of the two. A passive manager’s investment in equities mimics a stock market index and the stock is held for as long as it is in the index. (It should be noted that passive management is increasingly being used for other investment types, including bonds, commodities and

hedge funds.) Such a strategy is based on the hypothesis that markets are efficient and

reflect all relevant information, and seeks to minimise fees. Active asset managers, be they managers of retail or of institutional funds, pursue classic “buy and hold” strategies on the basis of a company’s underlying performance and competitive advantage. They seek to identify stocks which are under priced by the market and buy them in the expectation that over time the return will be greater than the market as a whole. To quote one active manager “our investment timescales looks 5 years or more years ahead in line with a typical timescale of underlying beneficiaries”. For both active and passive managers, there will always be changes in holdings at the periphery as they buy or sell as clients change or to align holdings according to changes in an index for an index fund. These changes tend to be small but can distort turnover figures. In this context, the European Commission’s Green Paper on a Corporate Governance Framework argued that financial market outlooks have become more short-term in recent years. We agree that in recent years there has been a growth of alternative short-term strategies, for example, hedge funds and other investors pursuing “quantitative” strategies that depend on short-term market trends and which involve turning over portfolios more rapidly. It is important that there is a diversity of equity investors as it promotes market

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

5

liquidity and efficiency. However, we do not agree that there had been any fundamental change in the way in which the assets of long-term investors are managed. The evidence we offered the Commission is set out in Annex 2. We will be submitting further evidence on asset managers’ timescales under separate cover. 2. How to ensure that shareholders and their agents give sufficient emphasis to

the underlying competitive strengths of the individual companies in which they invest. Evidence is invited on: a. how equity analysts and asset managers assess the competitive

advantages of companies.

Equity analysts analyse companies that make up the market and advise clients on the prospects for those businesses and industries over the short, medium and long-term. They work on both the buy and sell sides. The sell-side analyst discusses his recommendations with buy-side analysts who represent asset managers who, as outlined in 1d above, invest client funds. A company’s competitive advantage determines its future value and both equity analysts and asset managers review companies’ preliminary announcements and annual reports and accounts, and attend the twice yearly road shows in order to determine a company's value and potential. They also meet with company management to gain a better insight into the company's prospects and its management’s effectiveness. An asset manager will also study analysts’ reports.

b. the extent to which trading on equity markets is guided by analysis of

underlying corporate performance, and the extent to which it is driven by analysis of short-term market trends;

A distinction should be drawn between those who mainly trade shares (for example, banks and other proprietary traders) and those, like asset managers, that invest. Proprietary and principle traders that buy or sell equities with their own capital, including hedge funds and those with high portfolio turnover such as “high frequency traders”, tend to be driven by short-term market trends and turn their portfolios over rapidly. They will not tend to analyse underlying performance. Those that invest also buy and sell equities but tend to hold them for the long-term based on their analysis of a company’s prospects and underlying performance. An asset manager does not have an incentive to trade equities as the associated costs would reduce investment performance (and hence the manager’s competitiveness in attracting clients) and revenues due to the ad valorem nature of fees – see 6d.

c. how have technological advances such as automated trading affected investment decisions in equity markets;

Technological advances facilitate high-frequency trading where computer programs and specialised hardware enable traders to hold short-term positions in equities, options, futures, ETFs, currencies, and other financial instruments that can be traded electronically. Such traders compete with each other for small, consistent profits on the basis of speed and do not interact with company management. Such strategies can potentially impact market

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

6

volatility and are something that long-term investors have to deal with. For investors, technological advances such as CREST have speeded up the process of effecting investment decisions but have not impacted the decisions themselves.

d. whether corporate managers feel able to communicate effectively about issues related to the competitive position of their businesses.

Corporate managers communicate issues on the competitive position of their businesses through discussions with investors, twice yearly road shows, preliminary announcements, and annual reports and accounts. In the main this works well except that we are aware that at times companies may feel inhibited about communicating matters that are commercially sensitive. In addition, on its own, frequently the annual report and accounts does not provide adequate qualitative information and is issued too late after the end of the company’s financial year. (We are aware that HSBC issues its preliminary announcement and annual accounts at the same time. This should be commended.) Nevertheless reliable and clear annual reports and accounts that are subject to a quality audit have an important confirmatory role and are essential to maintaining confidence in the markets. They also have a vital role to play in helping investors understand what a company’s management has done with the resources entrusted to it, i.e. its stewardship. 3. Whether the current functioning of equity markets gives sufficient

encouragement to boards to focus on the long term development of their business. Evidence is invited on: a. whether changes in reporting obligations have influenced the perspectives

and timescales of managers and boards, and whether these changes in perspectives and timescales help or hinder long-term decision making;

We understand that US research highlights that quarterly reports encourage short-term thinking and discourage long-term investment on the basis they could impact quarterly performance. Over time, this erodes value creation and economic performance. Currently in the UK public companies are required to issue two interim management statements, as well as half-year and full year accounts. Whilst this is less onerous than quarterly reporting, we feared it was a step in this direction and we understand that a number of listed companies provide quarterly updates on performance. Provided companies are required to disclose promptly new developments, long-term investors do not need reports more frequently than every six months. We, therefore, welcome the recent proposal for an amended Directive on Transparency Obligations which removes requirements for interim management statements.

b. how the perspectives of managers and boards vary between listed

companies, companies whose equities are traded on AIM and PLUS;

We do not believe the perspectives of managers and boards necessarily vary between those companies that are listed and those whose equities are traded on AIM and PLUS. The exception is that it can be more difficult for smaller companies to attract the attention of the market as a whole as it tends to be those investors with larger stakes that take an interest.

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

7

c. whether publicly traded companies pay too much attention (or feel obliged to pay too much attention) to short-term fluctuations in their share prices;

In order to benefit their company’s shareholders, employees and other stakeholders, it is important that boards plan for the long-term as is recognised in Section 172 of the Companies Act 2006, which introduced a statutory duty for a company’s directors to act to promote the long-term success of the company. In this context, an important driver of company and board behaviours is the incentive structure for executives. Investors want their investee companies to have remuneration policies that are aligned with their interests such that they promote long-term value creation, take account of the fact that effecting change to a company’s strategy takes time, and mirror a company’s development cycle. The time frames involved will largely depend on the industry in which the company operates, but few are short-term - for example, in the pharmaceutical industry a particular drug’s development can take over ten years from the initial research. Nevertheless, remuneration structures are frequently based on short-term earnings and share price. Even long-term incentive plans that should reward successful implementation of a strategy and the creation of long-tem value rarely extend beyond three years. Also, benchmarking executive remuneration to the size of the business creates a motive to acquire businesses to boost directors’ earnings. There are a number of examples of such acquisitions which in the long-run were seen to destroy value. The short tenure of certain executives can compound this. A former President of the Chartered Institute of Personnel and Development recently stated that the average Chief Executive Officer only serves around four and a half years2 which often is not long enough to see the rewards from an investment. Also in the banking sector this was exacerbated by accounting requirements that allowed changes in the fair value of assets to be taken to earnings even when the holdings concerned were so large that they could not have been realised at those values. These unrealized gains were used as a basis for performance-related remuneration of both boards and bankers even though those gains never resulted in cash flows. This disconnect between earnings and value creation had to be subsequently reversed when the assumptions proved to be erroneous with consequences for the banks concerned. Other authors have argued that a focus on earnings has given wrong incentives to management and that alternative metrics should be considered3. In conclusion, directors’ remuneration structures can mean that they are incentivised to maximise share price or earnings in a short-time frame at a cost to long-term viability. We will be commenting on remuneration further in our response to the BIS Discussion Paper on Executive Remuneration. However, the link between pay and long-term performance and the alignment of investors’ interests are no longer always apparent. We consider the time horizons over which management is incentivised needs to be addressed, for example, executives should have to hold shares even after they leave the company concerned.

2 http://www.randstadinterimexecutives.co.uk/about-us/news-and-events/average-tenure-of-ceos-is-45-years-

expert-says

3 https://secure.cfauk.org/assets/2162/CFAUKDBIS_Long_Term_responseSENT.pdf

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

8

d. whether companies feel that their engagement with fund managers and analysts is properly focused on the competitive capabilities of the business.

Point d. is for companies to answer. In this context, in our response to question 8b, we explain the responsilbities for engagement within fund management organisations. 4. Whether Government policies directly relevant to individual quoted

companies (such as regulation and procurement) sufficiently encourage boards to focus on the long term development of their businesses. Evidence is invited on:

a. whether government policies encourage undue focus on cost cutting, or

otherwise damage the ability of firms to engage in long-term investment and the building of sustainable competitive advantage ;

b. whether government policies aimed at facilitating long-term investment by companies have been effective and whether there are other ways Government could support long-term business growth.

Although question 4 is for companies to answer, a lack of stability in Government policies can have unintended consequences when it comes to long-term investment and growth. In particular, policies which tend to follow the five year electoral cycle can potentially impair long term planning. 5. Whether Government policies directly relevant to institutional shareholders

and fund managers promote long-term time horizons and effective collective engagement. Evidence is invited on:

a. whether pension regulation, insurance regulation, supervision of

charitable endowments and regulatory requirements for asset managers lead to excessive emphasis on benchmarking and on short-term performance measurement;

The principal impact of regulation has been to drive long-term investors into fixed income investments at the expense of equities. This may damage long-term returns and in applying pressure at times of financial stress, as happened in 2003, can drive participants to standard types of behaviour such that risk is not diversified and equities are sold at a time when the risk premium is attractive depriving beneficiaries of the ensuing return. It can also impact investors’ ability to engage in that holders of bonds and other assets do not have the same rights as holders of equities. We set out below specific examples of the regulations concerned.

• Solvency requirements have impacted investors, particularly insurers, by obliging them

to shift from being long-term owners of shares to holding bonds and other assets.

Pension fund accounting requires pension scheme assets to be marked to market, whilst liabilities are discounted using AA-rated bond yields. This caused pension schemes to move out of equities into bonds to better match their liabilities.

• The UK tax system acts as a disincentive to invest in shares. Thus with an investment

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

9

in shares there is stamp duty on purchase, the dividend income is taxed and paid out of profits that have been subject to corporation tax, and there is tax on any gain on sale, whilst bonds and cash are only subject to income tax. In particular, stamp duty on the purchase of shares increases the cost of capital and reduces returns to investors. We would welcome its abolition.

b. whether the broader regulation of equity markets has an impact on the

investment timescales of market participants;

See a. above. In so far as asset managers are concerned, investment timescales are impacted by client needs as opposed to the broader regulation of the markets. The latter is more likely to give rise to cost issues and impact transaction based participants as opposed to the investment timescales of long-term investors. In this context, some have suggested that longer holding periods should be encouraged by giving preferential rights to those that hold shares for the longer term. However, there are significant practical issues in differentiating shares. For example, some investors (for example, index funds) are by definition long-term but on their initial investment would be treated as short-term. Nor is it clear why an active investor when it first makes a considered decision to take a position in a company– which could be the outcome of prolonged analysis and research – should be disadvantaged; this would almost certainly create market distortion. It is also unclear how changes in an existing holding would be treated; if an existing long-term holder’s new purchases were treated the same as their existing holdings, it would be very easy for any investor to secure better rights by taking very small holdings in many companies. Whilst we do not consider there need to be incentives to hold equities long- term, steps could be taken to encourage more investment in equities in that the UK tax regime acts as a disincentive, as noted in 5a. above.

c. whether the regulation of contact between companies and investors is an obstacle to effective engagement

UK asset managers typically have relatively small holdings, particularly in larger companies. Cooperation between investors, for example, in discussing the exercise of voting rights, in attending company sponsored presentations, can be vital to ensure engagement is effective. There are, however, concerns that acting collectively with like-minded investors to bring pressure to bear on management could trigger issues of insider trading, changes of control, and industry collusion and “the concert party” rules. There are dealing restrictions on investors that have inside information on a company's trading and performance. (In 2007 the Financial Services Authority clarified that this would apply should an investor trade on the basis of another’s strategy or if several acted together with a view to avoiding disclosures which would be necessary were the shares to be acquired by a single entity4.) Other uncertainties arise from the mandatory bid rules in the UK’s Takeover Panel’s Code and from requirements implementing the Acquisitions Directive

– see Annex 4 – on which FSA issued guidance5 and the Takeover Panel a Practice

Statement6.

4 http://www.fsa.gov.uk/pubs/newsletters/mw_newsletter20.pdf.

5 https://fsahandbook.info/FSA/html/handbook/SUP/11/Annex6G

6 http://www.thetakeoverpanel.org.uk/wp-content/uploads/2008/11/ps26.pdf

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

10

However, some of our members have confirmed that they still have concerns and that the Takeover Panel’s Practice Statement will not necessarily change behaviour in that uncertainty still remains. Also, following Sir David Walker’s review of corporate governance, the FSA wrote to the then Institutional Shareholders’ Committee in 2009, clarifying that there was nothing under FSA rules that would prevent investors discussing matters when it is for a legitimate purpose7. Whilst this was helpful, investors’ ad hoc discussions or understandings often go further than simply to quote “promoting generally accepted principles of good corporate governance” and certain of our members remain concerned that normal engagement activities could draw them into the Market Abuse regime. 6. Whether the current legal duties and responsibilities of asset owners and

fund managers, and the fee and pay structures in the investment chain, are consistent with these long-term objectives. Evidence is invited on: a. whether there is a more rapid turnover of asset managers and whether

this makes it more difficult for these managers to take a long term view of the companies in which they invest;

Our members do not consider that there is a more rapid turnover of asset managers which has impacted their ability to take a long term view. It is sometimes suggested that quarterly monitoring of performance by pension trustees in particular puts pressure on investment managers to target short-term returns for fear that they will lose the business because of a short period of underperformance. In order to test this suggestion, in 2004 we sponsored a survey jointly with the National Association of Pension Funds on the length of manager mandates8. This showed that, while fund managers reported feeling under some pressure from pension funds to deliver performance in the short-term, they typically retained mandates for five years or more. While performance will naturally be monitored regularly – typically quarterly – it would be very rare for mandates to be terminated on the basis of short-term underperformance. We do not believe that this has changed.

b. how individual asset managers are rewarded, and their performance

measured, and whether this gives insufficient incentive for them to take a long term view of the companies in which they invest;

As regards the incentives that operate, an individual portfolio manager’s performance is assessed on a medium to long-term basis, with other factors such as client satisfaction, attitude to risk, and the extent to which the employee is a team player taken into account. For example, for an individual fund managers' remuneration, the basic/fixed part is around 30 to 40% of the total and the performance part is around 60 to 70%, of which a significant amount is deferred over two to four years. To quote various asset managers:

“[Our] remuneration policy is team based and 75-80% of bonuses is paid in shares and has a three year vesting period. There is therefore no incentive to focus on one year’s performance.”

7 http://www.fsa.gov.uk/pubs/other/shareholder_engagement.pdf

8 http://www.investmentuk.org/assets/files/press/2004/20040913-01.pdf.

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

11

“[We] are increasingly charging performance fees, which are based on at least a year-on-year performance. Remuneration of individual fund managers is based on a mix of team, fund and individual performance (roughly a third each) and no changes have recently been made to this policy.”

“There is no linkage with fees and short termism if they are calculated on an ad valorem basis. [It] does have some funds with performance fees which are calculated each year. Where there has been some underperformance however the fund has to get back to its starting position before any subsequent outperformance can be rewarded. [It] believes this aligns them with the client and as they are building a long term relationship does not lead to taking risks in the short term.”

“[We] have no remuneration structures whether for managers or the company, which incentivize an increased turnover of securities.”

“[Our] individual asset managers have their remuneration linked to 1 and 3 year performance cycles.”

While the level of fees has an impact on performance, individuals are paid by the firm, not by the client, so that decisions about an individual’s remuneration do not affect the cost to clients. In any event, the remuneration of individual fund managers is being addressed through the implementation of the Capital Requirements Directive, and its application via MiFID and we do not believe there is a case for further regulation.

c. whether there are agency problems in the objectives and operations of asset managers that may be deleterious to the interests of the corporate sector or savers;

We often hear commentators referring to problems due to the agency nature of investment management but do not understand what this is alluding to. First, we believe that much of the academic research in this area is flawed and fails to recognise the difference between those that act as principal, such as banks and proprietary traders and asset managers that act as agents for clients. (We explained this in our answer to question 13 of the EU Commission’s Green Paper9.) As agent for its clients, an asset manager invests on behalf of clients in corporates which in turn seek to generate value that exceed the cost of capital. The client’s assets are safeguarded and remain the client’s and the manager, in both the retail and institutional markets, is generally remunerated on the basis of an ad valorem fee, representing a fixed percentage of assets managed. The manager’s incentives are, first, to retain the business long-term and, second, to increase the value of the client’s assets over time either as the client gives him more money to manage or if the value of the portfolio increases. The manager’s interests are aligned with those of both its clients and its investee companies – the better the companies perform the better returns clients or savers achieve which in turn means that the managers are better remunerated. Particularly in the institutional market, these ad valorem fees may be supplemented by performance fees. These will be negotiated and agreed in advance with clients and their advisers. We have spoken to a number of our member firms about the type of performance fees they agree with clients. Typically these will be based on rolling three-year performance figures; in some cases any period of underperformance must be followed by fully making up the “lost” return before any performance fee can be payable. An asset management firm

9 http://www.investmentfunds.org.uk/assets/files/consultations/2011/20110722-eugreenpaper.pdf

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

12

does not have an incentive to over-trade portfolios, because the associated costs would simply reduce investment performance (and hence a firm’s competitiveness in attracting clients) and its revenues, because of the ad valorem nature of fees. This is the reverse of the situation for banking intermediaries, where the bank in effect earns revenue by reducing the client’s return – see d. below.

d. how other intermediaries and market participants are remunerated and what impact this has on their incentives and those of their clients.

There are many participants in capital markets, including banks, brokers, market makers, other proprietary traders, hedge funds and asset managers. All have different business models and incentives. As noted above, a distinction should be drawn between those that act as principal and those that act as agents for clients. For years investors have been concerned about the leak of value to intermediaries, particularly the sell-side investment banks, on which they have little or no say. Sell-side professionals issue, recommend, trade and "sell" securities for buy-side investors, asset managers, to "buy". Unlike asset managers they are not remunerated through a growth in the value of investments, but by charging clients fees and commissions for services. The relationship is transactional and is a zero-sum game, i.e. one participant's gain or loss is exactly balanced by the losses or gains of the other participant. Thus the more the bank makes, the higher the cost to the client. The bank is, therefore, motivated to trade frequently and to persuade clients to do so as well. In practice UK listed companies also retain a corporate broker, now mainly integrated into the investment banks, to provide advice. They advise on mergers and acquisitions where a significant percentage of the fees paid are fixed and contingent upon the deal being completed, thereby creating incentives to ensure the merger is completed. As noted in 3c. there are a number of examples of acquisitions which in the long-run destroyed rather than maximised value. The broker also helps the company raise capital by underwriting and/or acting as their

agent in the issuance of securities. Typically, the broker’s parent, the investment bank,

uses its own balance sheet and acts as the primary underwriter. This creates an inherent conflict and a comparison of underwriting fees paid by a selection of companies that raised capital via rights issues to total directors’ remuneration in the year of issue showed that underwriting fees were ten times the reward for executive directors. The Institutional Investor Committee issued a report in 2010, Rights Issue Fees Inquiry, and made a number of recommendations on this process, which we would like to see promulgated10. We would also welcome more fees being deferred and payment being dependent on the transaction securing the long-term success of the company as opposed to simply its completion. 7. Whether there is sufficient transparency in the activities of fund managers,

clients and their advisors, and companies themselves, and in the relationships between them. Evidence is invited on:

a. whether the existing rules on disclosure of material stakes are excessive

or inadequate;

10

http://www.iicomm.org/docs/rifireport.pdf

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

13

The disclosure of major shareholdings is an important component of company law which provides information on shareholders accumulating shares in circumstances where they may seek to influence or control a company. It allows investors to see movements in shares prior to a particular investor obtaining possible influence over the company.

IMA does not consider that the existing rules on disclosure of material stakes and the Companies Act thresholds (i.e. 3% and 1% thereafter) are excessive or inadequate. However, these thresholds only apply to UK incorporated issuers whose home state is the UK and not to non-UK issuers (third country issuers) whose home state is the UK. Some IMA members considered that the FSA should have applied the Companies Act requirements to non-UK issuers for whom the FSA is the home competent authority as opposed to limiting the requirements to the TOD minimum. They do not consider that a two-tier system, which depends on where the issuer is incorporated, is sensible. Other members are more concerned with the general lack of equivalence in the context of notifications in respect of third country issuers in that this has given additional burden for major holders in having to make two sets of notifications.

b. whether asset managers should be subject to more extensive disclosure requirements, e.g. of costs and remuneration structures;

IMA does not consider that it is necessary for asset managers to be subject to more extensive disclosure requirements around costs and remuneration structures. Investment managers owe their duty to their clients. Those clients may be institutions which have given specific mandates which the manager must adhere to: it is for the client to specify how it wishes the money to be managed and what reporting requirements to put in place. Managers are fully transparent with their clients and will provide them with the information they request. For retail funds, there is not of course the same dialogue between manager and client. But authorised retail funds are subject to strict regulatory rules requiring them to be transparent about charges, pricing, portfolio composition and other matters. For example, under requirements that implement the relevant European Directives asset managers that manage Undertakings for Collective Investment in Transferable Securities (UCITS) have to produce Key Investor Information Documents which describe a fund’s objectives, its risks & rewards and also any charges; these are presented in a prescribed way to enable investors make informed decisions. Lastly, it also needs to be considered that more transparency can often give rise to unintended consequences and has effectively accelerated remuneration in the corporate sector.

c. whether the growth of investment consultants has encouraged or discouraged engagement by share owners with companies;

The majority of Requests for Proposals prepared by investment consultants now contain questions on the extent of the asset managers’ engagement with companies and thus have tended to encourage engagement.

d. whether the overall costs of intermediation are understood by beneficiaries, and are proportionate to the value of the services provided;

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

14

Question 7d is for beneficiaries to answer as to whether costs are understood. Asset managers are transparent as to their costs and the Retail Distribution Review which is part of the FSA’s consumer protection strategy should provide transparency as it addresses intermediary costs.

e. whether investors have sufficient information to understand the investment approaches of asset managers and to judge whether they are aligned with their investment objectives and timescales.

Institutional investors will negotiate the terms of their mandates with asset managers and ensure that these are aligned with their investment objectives and timescales. In particular, pension schemes will draw up a Statement of Investment Principles (SIP) which sets out the principles on how investment decisions should be made which any appointed asset manager must follow. In the case of a retail investor that uses an Independent Financial Adviser it is reasonable for the investor to rely upon the adviser’s skill, qualifications and competence to address suitability and alignment with the investor’s objectives and timescales. Where the investor does not have such an adviser, he would need to rely on published information, such as that in the Key Investor Information Document and any other material that complements it.

As regards engagement, it is now an FSA requirement that investment managers disclose on their websites their commitment to the UK Stewardship Code or their alternative business model11. This ensures that those that appoint investment managers are aware of how a manager exercises its stewardship responsibilities, if any. The Code also expects those that commit to it to report to their clients on how they have exercised their responsibilities and also to have a public policy on voting disclosure and either disclose publicly how they have voted or where they do not explain their reasons – a “comply or explain” approach. 8. The quality of engagement between institutional investors and fund

managers and UK quoted companies, and the importance attached to such engagement, building on the success of the Stewardship Code. Evidence is invited on:

a. whether the measures taken to stimulate engagement by investors with

companies have been sufficiently effective; The UK Stewardship Code sets out good practice for engagement. In May this year IMA published a report on “Monitoring Adherence to the Stewardship Code” 12 which looked at the activities that support institutional investors’ commitment to the Code. It was compiled from the answers of 41 Asset Managers, seven Asset Owners and two Proxy Advisors to a questionnaire which covered the period to 30 September 2010. The Managers that responded managed £590 billion of UK equities representing 31% of the UK market and the Owners owned £15 billion. (Proxy advisors do not hold equities.)

11

The Stewardship Code is now the responsibility of the FRC which is listing those managers that

commit to the Code on its web site with a link to the manager’s policy statement. 12

http://www.investmentfunds.org.uk/research/stewardship-survey

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

15

The survey demonstrated widespread adherence to the Code:

over 90% of major institutional investors voted all or the great majority of their shares in UK companies; nearly two thirds published their voting records;

at the time of the survey, 43 out of 50 respondents had published a statement on adherence to the Code, and another six did so subsequently; and

over 1,300 people focusing on stewardship activities are employed by 43 of the respondents to the survey.

The report included case studies showing how investors dealt with six controversial issues, ranging from remuneration, Board composition, environmental concerns, mergers and financing. All cases saw high levels of engagement with companies, which took a variety of forms, including meetings at senior Board level, consultation with other investors, and, in one case site visits. Outcomes varied from changes in Board policy to improved transparency; in one instance the outcome was reluctant acquiescence on the grounds that alternative courses were less satisfactory. In over 80 per cent of cases, investors believed they achieved their objectives in engaging on the issue. IMA recently issued the questionnaire for the 2011 survey to the 175 signatories that had signed up as at 30 September 2011 and aims to publish in March 2012. At this stage it is too early to comment on the progress and how effective the Stewardship Code has been. More time should be given for this initiative to take effect.

b. whether the corporate governance activities of asset management businesses are sufficiently integrated with the decisions of fund managers.

Corporate governance activities, in so far as this refers to engagement, may take place at any level within an asset management business – they may involve individual fund managers, analysts and corporate governance staff, and the corporates may include both executive and non-executive directors, and other management and personnel. In the main, matters relating to strategy and performance are handled by the portfolio managers/analysts but, due to the specialist knowledge required, dedicated stewardship specialists handle aspects such as corporate governance and socially responsible investment. As may be implied by the question, this dual approach has given rise to questions as to how stewardship is integrated into the investment process. More recently, the Guidance to Principle 1 of the Stewardship Code expects institutional investors to disclose their internal arrangements on how stewardship is integrated into the wider investment process. This was also one of the areas explored in IMA’s survey on Monitoring Adherence to the Code. To quote the final report: “every respondent that adopts such an approach integrates stewardship in one or more ways. For example, those involved in the investment process may set and/or approve the stewardship policy, make the final voting decision in a controversial situation, attend meetings with investee companies with the stewardship specialists and/or simply meet the stewardship specialists in house”. The full extract from the final report is set out in Annex 3.

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

16

9. The impact of greater fragmentation and internationalisation of UK share ownership, and other developments in global equity markets, on the quality of engagement between shareholders and quoted companies.

Evidence is invited on:

a. what has been the effect of the internationalisation of UK equity markets

on the priorities of companies and fund managers;

Around twenty years ago UK insurance funds and pension schemes were typically the major long-term owners of UK equities with exposures approaching 60% of the UK equity market. Since the mid-1990s the proportion of UK equities owned by UK institutional shareholders has declined from 60 to 40%. There are several reasons for this. First, there has been a trend worldwide towards investing in international equities and not just domestic shares. Secondly, the equity allocations of UK pension funds and life insurers have declined, reflecting the changing nature of liabilities and the impact of accounting and solvency based regulation – see 5a. At the same time while there has been strong growth in UK mutual fund holdings of UK equities, this has not been sufficient to counteract these trends. The internationalisation of the UK equity markets has been generally good for corporates in that it has increased the available capital and reduced its cost. The recent influx of overseas mining companies seeking a listing is testament to London’s success in attracting major issuers to London. However, our members are concerned by recent examples where the minimum free-float requirements were waived prior to listing and have not been reinstated. This raises questions to the extent to which the minority shareholders will be afforded the same standard of protection expected of a premium listing. It is vital that the quality of a UK listing is maintained and particularly the integrity of the list in that this is very often a passport to being included in the main FTSE UK Indices and accessing the funds of the large passive asset managers.

b. whether the growth in overseas ownership of UK equities, and in the overseas activities of UK listed companies, has affected engagement between UK investment institutions and UK companies.

It can be more difficult for overseas shareholders to engage with UK companies which, given their increasing presence on UK share registers, can further limit what engagement can achieve. Moreover, whilst this may mean that the holdings of UK investors represent a larger proportion of the engaged share base such that they may be able to exert more influence on the UK companies in which they invest, we understand that certain companies can interpret the silence of overseas investors as support. 10. Likely trends in international investment and in the international regulatory

framework, and their possible long term impact on UK equity markets and UK business.

Evidence is invited on:

a. how UK asset managers, and UK companies, expect the pressures on them to change with further internationalisation of equity investment;

ANNEX 1 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED

17

b. whether recent or planned regulatory actions by authorities outside the

UK, and particularly regulatory policy developments at EU level, will affect engagement between asset managers and the companies in which they invest, and the ability of companies to respond to that engagement.

The EU is yet to publish proposals in relation to its review of the corporate governance framework. One of our main concerns is that the EU’s Green Paper implied that the regime of “comply or explain” corporate governance statements is not effective and suggested that they should become regulated information. We do not agree with this. Corporate governance is about behaviours which cannot be addressed in legislation. Legislative measures can result in standardised disclosures to ensure compliance rather than a sound corporate governance culture and lack flexibility such that a company would not be able to tailor them to its own circumstances. They would also undermine the role played by shareholders who, as the providers of the risk capital and bearers of the residual risk, are best placed to monitor corporate governance arrangements, engage with companies and hold their boards to account. As noted in 7 and 8 above, the UK gave new impetus to engagement last year and issued a code of best practice for institutional investors and the UK regulator now requires investment managers to disclose publicly their commitment to this code or their alternative model. IMA published a report which clearly demonstrated investors’ commitment to the code. EFAMA also published its own code for investor engagement which is largely based on the UK’s code. This framework of codes for both companies and investors is relatively new and should be given time to take effect.

ANNEX 2 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED EXTRACT FROM IMA’S SUBMISSION TO THE EUROPEAN COMMISSION’S GREEN PAPER – THE EU CORPORATE GOVERNANCE FRAMEWORK

18

“No evidence of increased turnover or “churning” of client portfolios by asset managers The Paper also refers to increased turnover in capital markets. It notes, after Woolley and also Haldane at the Bank of England, turnover of some 150% of market capitalisation, and concludes that this implies average holding periods of eight months. The implication appears to be that there is a concern that fund managers are unnecessarily “churning” client portfolios. We believe this is incorrect. As noted above, there are many participants in capital markets. They include a number with high portfolio turnover, such as “high frequency traders” and some hedge funds. We agree that such activity has been increasing in recent years and believe that the observed increase in market turnover is the result of this. The capital markets research consultancy Tabb Group produced the following breakdown earlier this year of turnover in the UK equity market:

Hedge funds 37% High frequency traders 28% Investment bank proprietary trading 7% Retail investors 4% Long-only funds 24%

Source: Tabb Group “Breaking Down the UK Equity Market”, January 201113 This shows that high quoted turnover rates are largely the result of activities by players other than institutional asset managers. This group however manages the overwhelming bulk of institutional assets: in the UK the “traditional” asset management industry is about twenty times the size of the hedge fund industry. Nor can we find evidence of investment managers holding the shares of individual companies for shorter periods. We have asked our members for data about portfolio turnover within their funds. Most have reported typical holding periods for stocks of at least four years and no secular rise in turnover over time, though turnover does increase temporarily during periods of market turmoil, such as 2008-09. Please see Annex 2 for turnover figures for two managers’ funds. We have further corroborated this by examining receipts from stamp duty reserve tax in the UK. This is a 0.5% tax payable when shares are bought, although banks enjoy an exemption as market makers. Total receipts from this tax are published by Her Majesty’s Revenue & Customs. We have grossed-up these figures to reach implied total turnover and then compared them with average total market capitalisation for the relevant tax year. This gives what we believe to be a reasonable proxy for average turnover across the market. The results are as follows:

13

This research is proprietary, but Tabb have confirmed that they would be pleased to share it if that were

helpful.

ANNEX 2 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED EXTRACT FROM IMA’S SUBMISSION TO THE EUROPEAN COMMISSION’S GREEN PAPER – THE EU CORPORATE GOVERNANCE FRAMEWORK

19

Tax year Turnover as % of

market cap Notional average

holding period (months)

2001-2 36% 33 2002-3 42% 29 2003-4 40% 30 2004-5 36% 33 2005-6 30% 40 2006-7 26% 46 2007-8 26% 46 2008-9 35% 34 2009-10 27% 44 Source: UK National Statistics, London Stock Exchange and IMA calculations It is clear from this table that there has not been any trend over the last ten years towards higher turnover, although it was unsurprisingly raised in the wake of the bursting of the dot.com bubble in the early years of the last decade and during the credit crisis in 2008-09. To the extent that these calculations include trading entities which pay stamp duty, they may overstate turnover by asset managers. Moreover much turnover may be the result of factors such as managers having to change their holdings in response to flows in and out from clients, or decisions to increase or reduce exposure to a stock while remaining a shareholder. It follows that the periods for which even active managers are likely to remain shareholders in individual companies are likely to be significantly longer than the notional holding periods in the table. We conclude that we have yet to see evidence of what the Paper describes as inappropriate short term behaviour by investors generally. Many managers of retail funds and of institutional money continue to pursue classic “buy and hold” strategies. Indeed this is the essence of “active” fund management – the manager seeks to identify stocks which are under-priced by the market and buys them in the expectation that over time the return will be greater than that of the market as a whole. This is wholly consistent with what one would expect to find given the incentives against portfolio “churning” by asset managers described above.”

ANNEX 3 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED CONCERNS ABOUT ACTING IN CONCERT

20

The uncertainties around collective engagement arise first, from the mandatory bid rules in the UK’s Takeover Panel’s Code and secondly, from the requirements implementing the Acquisitions Directive. Rule 9 of the UK’s Takeover Code requires that if a person and any person acting in concert with that person acquire, whether by a series of transactions over a period of time or not, an interest in shares which together exceed 30 per cent, a mandatory offer has to be made for all the shares of the company. The underlying philosophy is that if control of a company passes into one hand, all shareholders should have the chance to dispose of their shares at the highest price paid by the new controller. First, they may not wish to remain in the company under a new controller. Secondly, as all shareholders (not just the old controller) should share the premium paid for someone acquiring control the buyer has to offer all shareholders the highest price he paid for his shares. As a result of note 2 to Rule 9.1 of the Code14, investors are concerned that if they act collectively such that the combined ownership of the parties “acting in concert” exceeds the threshold of 30 per cent and they seek to change board representation, they would have to make a mandatory bid for the company. Moreover, we understand that the parties involved would be jointly and severally liable – each could be left with the responsibility to make a full bid. There are also concerns that there could be fines and negative reputational impact arising from regulatory proceedings if they were made public. It would be helpful if the Panel could address investors’ concerns in this area. Secondly, there are issues as to whether collective shareholder action in relation to banks and other financial institutions are caught by the Acquisitions Directive which had to be implemented across the EU by 21 March 2009. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2007:247:0001:0016:EN:PDF In summary, the Directive reforms the provisions of the EU sectoral directives requiring regulatory approval before an investor acquires a direct or indirect "qualifying holding" in a bank, investment firm or insurance company (broadly speaking a holding of 10 per cent or more of the shares or voting rights - the UK transposition uses the term "control" instead of qualifying holding but is otherwise substantially the same). The amended sectoral Directive provisions state: "Member States shall require any natural or legal person or such persons acting in concert (hereinafter referred to as the proposed acquirer), who have taken a decision either to acquire, directly or indirectly, a qualifying holding in [an insurance undertaking, credit

14 Note 2 to Rule 9.1 states: "The Panel does not normally regard the action of shareholders voting together on a particular resolution as action which of itself indicates that such parties are acting in concert. However, the Panel will normally presume shareholders who requisition or threaten to requisition the consideration of a board control-seeking proposal either at an annual general meeting or at an extraordinary general meeting, in each case together with their supporters as at the date of the requisition or threat, to be acting in concert with each other and with the proposed directors. Such parties will be presumed to have come into concert once an agreement or understanding is reached between them in respect of a board control-seeking proposal with the result that subsequent acquisitions of interests in shares by any member of the group could give rise to an offer obligation.”

ANNEX 3 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED CONCERNS ABOUT ACTING IN CONCERT

21

institution, etc.]...first to notify the competent authorities..." The Directive does not seek to define when a person is “acting in concert” for these

purposes. However, the EU Level 3 Committees have given guidance on this issue.

http://www.cesr-eu.org/data/document/08_543b.pdf. Appendix 1 states:

"In the particular context of Directive 2007/44/EC, persons are 'acting in concert' when each of them decides to exercise his rights linked to the shares he acquires in accordance with an explicit or implicit agreement made between them. Notification of the voting rights held collectively by these persons will have to be made to the competent authorities by each of the parties concerned or by one of these parties on behalf of the group of persons acting in concert."

There are concerns that this definition would affect investors’ ability to cooperate, as even an ad hoc agreement or understanding to vote together on any issue would appear to result in the parties being treated as “acting in concert”. We consider there is, therefore, a risk that they would need prior regulatory approval to enter into that agreement or understanding if their holdings together exceed 10 per cent (or if, while the agreement or understanding subsists, either of them acquires additional shares that takes their aggregate holding over this threshold).

We understand that the FSA’s current procedures do not envisage such notifications and approvals and there are concerns that the process to be followed and time taken would be an impediment to collective engagement. To address these concerns, it would be necessary to have some clarification of the EU Level 3 guidance that the Directive’s provisions are not triggered by an ad hoc agreement or understanding to vote together on a particular issue. In addition, because UK listed banks and insurers often also own regulated subsidiaries in other EU countries, the Level 3 Committees (or the European Commission) will need to confirm this understanding if shareholders are going to feel comfortable relying on any UK

specific guidance.

ANNEX 4 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED EXTRACT FROM REPORT, ADHERENCE TO THE FRC’s STEWARDSHIP CODE AT 30 SEPTEMBER 2010

22

“INTEGRATION INTO THE INVESTMENT PROCESS As noted, engagement on strategy and performance is often handled by the portfolio managers/analysts but stewardship specialists handle particular aspects such as corporate governance and socially responsible investment. At times this dual approach can give rise to questions as to whether those responsible for stewardship represent the views of the portfolio managers responsible for the investment. It can also be important to have an

integrated approach – according to one Asset Owner: “in our experience, stewardship is most effective when integrated into the investment process - where strategic, operational and financial

issues are considered alongside ESG [Environmental, Social and Governance] factors”. The Guidance to Principle 1 addresses this and states that the stewardship policy should disclose: “internal arrangements, including how stewardship is integrated with the wider investment process”. To determine what happens in practice, the 31 respondents (26 Asset Managers and five Asset Owners) where stewardship specialists have a key role were asked how they ensure integration into the investment process from a range of options - Table 6. Each Asset Manager selected at least one option and integrates stewardship into the investment process in some way. For more than 88 per cent (23 out of 26 Managers), those involved in the investment process set and/or approve the stewardship policy and for 76 per cent (20), they make the final voting decision in a controversial situation. Around 88 per cent indicated that stewardship specialists attend meetings with investee companies with portfolio managers/analysts, with the same percentage simply meeting in-house with portfolio managers/analysts. Of the five Asset Owners, two have in-house investment teams that make the final voting decision on controversial issues and one of these teams also sets and approves the stewardship policy. For these two Owners, the stewardship specialists also attend meetings with investee companies with the portfolio managers/analysts. Another Owner retains stewardship in house and meets with companies but delegates investment to external Asset Managers sometimes discussing stewardship with them. Two Owners outsource both investment and stewardship to external Asset Managers though one retains discretion over voting. It is clear from the comments received from Managers that the approach to integration varies.

“Portfolio managers meet regularly with the executive management of the companies that

we invest in and the stewardship specialists often attend such meetings. In addition, the

stewardship specialists arrange separate meetings to discuss governance issues, often with

non-executive directors or with the Company Secretary. We liaise with the portfolio

managers ahead of these meetings and often the portfolio managers will attend”.

“Stewardship specialists sit with the portfolio managers to facilitate formal and informal

dialogue on stewardship issues. Whenever possible, portfolio managers attend [our]

meetings with Chairman together with the stewardship specialists.”

ANNEX 4 IMA’s EVIDENCE IN RELATION TO THE QUESTIONS RAISED EXTRACT FROM REPORT, ADHERENCE TO THE FRC’s STEWARDSHIP CODE AT 30 SEPTEMBER 2010

23

“Provides a full ESG overlay service on its clients' investments and does not in itself

undertake any investment. However, [it] is pleased to interact with its' clients’ fund

managers and to provide input on ESG issues into their own company analysis as required”.

One formally combines its research and ESG (Environmental, Social and

Governance) teams and uses the joint analysis to rank stocks which is reflected in

investment decisions.

Another has a committee of ten portfolio managers/analysts and three stewardship

specialists that make decisions. For contentious issues, the portfolio

managers/analysts are consulted, and the issue may be escalated to the committee.

Table 6: Integration into the investment process

Asset Managers

Asset Owners

Investment process sets and/or approves stewardship policy

N/A

23

3

1

4 Investment process makes final decision on a controversial vote

N/A

20

6

2

3

Attend meetings with portfolio managers/analysts15: Always 2 -

Often 5 1 Sometimes 16 2

N/A 3 2 Regularly meet portfolio managers/analysts

N/A

23

3

3

2

Certain respondents not included in Tables 5 and 6, chose to comment. “[Our] responsible investment specialist sets the high level stewardship policies. Detailed

policies are set by [an overlay service provider] in consultation with [us]”.

“Stewardship activities are the responsibility of the portfolio managers/analysts ("Investment

Professionals") and are therefore fully integrated into our investment processes, rather than

being delegated to stewardship specialists.”

[It] is a shareholder body with an explicit mandate to sell the investments it has in the

banks. ....our overall success will be dependent on how we discharge our stewardship role

and on how we dispose of the investments. Stewardship is therefore integral to everything

that [it] does and so there is not a distinction between an investment team and a

stewardship team. It is the responsibility of everyone within the team including, crucially,

[its] Board of Directors which takes all strategic decisions”.

In addition, one Service Provider commented:”we do not invest client assets and therefore do not employ investment staff. ......Where there are particularly high-profile or controversial issues, or cases that fall outside our existing guidelines, we call an internal policy forum to discuss the issues and decide the voting position to adopt”.

15 Where respondents indicated more than one option, they were allocated to the least frequent

category.