seeing through the british pension system

7/31/2019 Seeing through the British pension system

1/30

12SEEING THROUGHBRITISH PENSIONSHOW TO INCREASE COSTTRANSPARENCY IN UKPENSION SCHEMESDAVID PITT-WATSON

HARI MANNJULY 2012

www.thersa.org


2/30


3/30

Contents 1

Contents

Acknowledgements 3

Tomorrows Investor: introducing ft- or-purposepensions to the UK 4

Introduction 6

Levels o transparency 9

Why dont customers demand change? 15

What in ormation should be provided? 18

Summary and recommendations 22

Appendices 24


4/30


5/30

Acknowledgements 3

Acknowledgements

The authors would like to thank the many people involved in this work.

In particular, the advisory board o Tomorrows Investor: its chair,Sir John Banham, and its members, Robin Ellison, Philip Goldenberg,Alwin Oerlemans, Matthew Taylor and Lindsay Thomas.

Thanks also to Pinsent Masons or its help in resolving legal issuesand APG, the leading Dutch pension und, or sponsoring the work.


6/30

Seeing through British pensions4

Tomorrows Investor:introducing ft- or-purpose pensionsto the UK

Three years ago, the RSA began investigating the e cacy o the UKinvestment system. A ter in-depth research, including the use o citizenjuries, we concluded that it was not t or purpose. Private pension savingin particular was ound to be patchy, costly and poorly structured.

However, we suggested that, with some modest changes, the privatepension system in Britain could be radically improved. This requirestwo things: rst, a system o low-cost, auto-enrolled pension provision;second, a recreation o collective pension structures which share risks andcan o er superior bene ts.

In line with these initial recommendations, we welcomed the govern-ments decision to adopt auto-enrolment, but noted the need to removearti cial restrictions it had placed on NEST, the de ault provider est-ablished to ensure the policy could work. Further, it should encourageother providers who would o er similar and transparent terms to NEST.We are there ore happy that the Parliamentary Work and PensionsCommittee is investigating the restrictions on NEST, and that whengiving evidence to the committee, the minister agreed that, without suchrestrictions, NEST would be more e ective and less costly. 1 We are alsopleased that two alternative low-cost providers have been established inthe UK, one o them a ormer sponsor o the RSAs work.

With regard to the second suggested change, the creation o collective

pensions, we are also pleased that the pensions minister has indicated hewould like to acilitate risk sharing in pensions 2 a central element o ourproposals. He has announced a consultation later this year to help achievethis. So, while we await action, the debate is moving in the right direction.

Throughout our work, we have noted the need or stakeholders toreach consensus on how best we provide pensions. So we are delighted thatthe National Association o Pension Funds (NAPF) and the TUC haveendorsed our proposals, and that we have also received help rom the CBI.We are encouraged that they are now working together to protect collective

1. See evidence at www.parliamentlive.tv/Main/ Player.aspx?meetingId=99682. See report o speech at http://ipe.com/news/webb-pledges-to- acilitate-pension-guaran-

tees-risk-sharing-in-dc_42567.php [requires ree registration to access]


7/30

Tomorrows Investor: introducing t- or-purpose pensions to the UK 5

pensions in the UK rom the unintended consequences o ill-judged regul-ations. 3 We would also like to thank the nancial press or the exposurethey have given to these issues o cost; not just the reports o the RSA butothers work as well. The tide o opinion is changing.

In this paper, we return to the question o pension charges. In partic-ular, how it is possible to in orm savers o precisely what charge is being

levied on their pension savings.

David Pitt-WatsonHarinder Mann

3. See www. t.com/cms/s/0/397de0a2-5574-11e1-9d95-00144 eabdc0.html#axzz1mTbjxzFF[requires ree registration to access]


8/30


Introduction

Our nancial system manages the savings o millions o peoples pen-sion unds and savings. Four years ago, the RSA began investigating theneeds o these citizen investors. Under the project umbrella TomorrowsInvestor, the work used citizen juries to probe areas where people elt im-provement was merited. Jury members concluded that they were unawareo pension charging structures and the amount o their investment that wasswallowed up in charges. It was this area they elt needed greatest re orm.

Since then, there has been a lively debate about pension charges,with ront page stories in national newspapers, and investigationscommissioned by parliamentary committees. But a conundrum remains.I this is such an important issue, why have market orces not broughtabout a re orm o the system? One reason is simply a lack o transparencyabout what the costs o pension savings are. Do we know the level o charges on our pensions pots?

In this paper, as part o the 2012 series o outputs rom theTomorrows Investor programme, we investigate the debate around costsand transparency in UK pension schemes. We discuss pension chargesand cost transparency in the UK. The ndings show that individualconsumers do not have a clear understanding o the charges being levied.Nor do small companies who, in good aith, are trying to o er decentpensions to their employees. Even more worrying, our research suggestspension providers are not reely giving the in ormation needed on costsand charges, or noting its signi cance or pension outcomes.

We ask why the market has ailed to deliver, and draw on behaviouraleconomics to o er some explanations as to why people do not shoparound to seek transparency and lower costs. We also examine andreview the most recent government investigations into costs and notethese have ailed to discover the level o pension charges, revealing insteadwidespread con usion about cost levels.

In the second part o the paper, we discuss a way orward or bringingtrust and transparency back to the British pension system. A case study o the Danish system provides an example o transparency and disclosure inpractice and we give an outline o its key eatures and principles. A simi-lar system is to be introduced in Holland.

Finally, we issue a call or action to bring together a change in the UKpension system on transparency and set out the steps needed to revivetrust in the system. In this context we welcome the recent consultation

rom an NAPF-led working group about how pension charges can bemade clear and comparable or employers, and we look ahead to howtransparency might also be achieved or individuals.

I this is such animportant issue,why have market

orces not brought about re orm o thesystem? One reasonis simply a lack o transparency about what the costs o

pension savings are


9/30

Introduction 7

What we know about costsIn previous reports, 4, 5 we have discussed the issue o why costs matter.This has o ten been overlooked because pension costs are expressed as anannual charge on the balance saved. However, the average time duringwhich money is held in pension savings might be 25 years. On that basis,a 1.5% charge per annum translates into 37.5% over the li etime o the

pension. When this was explained to our citizen jurors, they were aghast.They had not understood the mathematics o charges, nor had it beenexplained to them. This, more than anything, made them eel the systemwas not trustworthy. So it is perhaps worth taking an example o howcharges a ect returns.

Imagine that a 25-year-old decides to save or a pension, so they canretire at 65 and enjoy a pension or the next 20 years. They set aside 1,000each year, and raise that sum to cover infation, which is 3%. They receivea 6% return on their money. That means that, by the age o 65 i they haveno ees to pay, they will have a pension pot o 248,170. This in turn willcreate an infation protected pension o 16,080 or the next 20 years. Nowimagine that, each year, they need to pay 1.5% on the money they havesaved. How much will that reduce the value o the pension? The answer isthat it will be reduced to 9,900 each year. In this case, someone who paysno ees gets a 60% higher pension than someone who pays 1.5%.

What initially seems like a modest charge actually equates to avery large di erence in pension payment. Compounding o charges isone reason why citizen investors have such di culty in distinguishingbetween one pension and another. It is simply not made clear to themthat a pension provider with a charge o 0.5% will, all else being equal,provide a pension which is a third higher than one which charges 1.5%.

What we dont know about costsBut even i the calculation we have just described were made clear,consumers need to know about all charges. Right now they do not.

Savers are requently quoted the AMC , or annual managementcharge . This is what the und manager takes or their service every year,and is generally accepted as the key indicator o what a pension will cost.However, the und manager is not the only one who adds charges tothe und. Audit, custodial and other costs are o ten charged separately.Further, some unds place savers money with others. So the top undin which a saver invests may charge 0.5%, which will be disclosed as the

AMC, but then subcontracts some o the investment activity to anotherone which charges 1%, creating a total charge o 1.5%. In order to aggre-gate all these charges, pension providers are now being asked to providea total expense ratio (TER) ; that is, how much in total they charge annu-ally as a percentage o the unds they manage.

But there are still urther, hidden costs in managing a pension und.These costs include taxes, broker commissions and the di erentialbetween bid and o er prices. They also include elements such as stocklending ees, and a list o other costs that are associated with di erent

4. Tomorrows Investor: Building the con sensus for a Peoples Pension in Br itain ,

RSA, 20105. Tomorrows Investor: Pensions for the people: addressing the savings and investment

crisis in Britain , RSA, 2010


10/30


asset types and investment strategies. These costs are not declared, andwe have not been able to nd a pension provider willing to give us a ullbreakdown o them. As a result, no one can tell the ull costs associatedwith pension und investment.

In October 2011, the NAPF warned o the eye-watering complex-ity related to pension und charges. 6 But it does not have to be like this.

Across the North Sea, in Denmark, pensioners are provided with a clearunderstanding o the costs attributed to the und. The Danes go urtherin their disclosure on a variety o topics a ecting a unds per ormanceand ask pension providers to give a clear account o their position. Thistransparency on both the investment and operational side o the pensionmeans the charging structure and potential value o the uture pensionpot are clearly understood. Consumers can choose which pension pro-vider is likely to give them the better return and so allows market orcesto work e ectively.

These examples show that the problem can be addressed in the UK,but only i we see much greater transparency in our pension system.

6. Daily Telegraph , 19 October 2011, www.telegraph.co.uk/ nance/personal nance/pensions/8837642/Prudent-savers-hit-by-excessive-hidden- ees-on-pensions.html


11/30

Levels o transparency 9

Levels o transparency

Retail pensionsIn February 2012, we conducted a short survey o the in ormationgiven to individual, or retail consumers when purchasing, or planningto purchase, a pension. We contacted a sample o 23 pension providersin the UK and asked the ollowing questions:

What charges would be levied to the pension investment? What other charges would be levied apart rom the AMC and

administration charges? Are you able to con rm that no other urther charges are levied

to the account? Is it possible to get a breakdown o any charges?

This ollowed our 2011 research 7 when we asked Fellows o the RSAto ascertain rom pension providers what they were being charged ortheir pension in relation to costs. In both cases, we wanted to understandwhether the in ormation given on costs accurately refected the true posi-tion o the scheme.

Our fndingsO the 23 pension providers questioned, 21 responded by saying that theannual management charge and the administration costs were the onlycharges. O the two that did say there might be other possible chargesaccrued, they were unable at the time o publication to give a list o read-ily available charges. Yet we know that there are bound to be chargesbeyond these: or example, or trading or or stock lending.

It can be argued that trading and other costs add value to the saversport olio, and are there ore worth incurring; indeed, that a good und

manager will wish to incur them i it is in the savers interest. But regard-less o the merit, it does not mean the cost has disappeared. In a recentstudy or the Treasury, Christopher Sier and ormer hedge und managerDavid Norman sought to disentangle the ull costs o investing in com-pany shares. 8 They ound the TER using data rom a pro essional undratings agency. 9 They then looked at the average cost and requency

7. RSA Comment, David Pitt-Watson http://comment.rsablogs.org.uk/2011/01/05/tomor-rows-investor/

8. Sier and Norman, Compexity and overintermediation in UK equity fund management ,Nov 2011

9. Lipper, a Thomson Reuters company, which describes itsel on its website(www.lipperweb.com) as a global leader in supplying mutual und in ormation and

und ratings, und analytical tools and und commentary.


12/30


o trading shares. They reckoned the TER was 1.7%, and had been risingover previous years, and that trading costs were 1.4%, making a totalcharge o 3.1%. This was or an equity und but i such a charge wasapplied over the 25-year li e o a pension und, more than hal the poten-tial pension bene t would disappear.

In Table 1 we illustrate our survey results. As mentioned, even when

we probed, only two out o 23 respondents admitted there might be trad-ing and other costs. Fees or similar services varied hugely: rom 0.1% to1.5%, a 15- old di erence. And all the ees are lower than the TER in thesurvey used by Sier and Norman.

For this reason we re-contacted some o the providers, to probe moredeeply into the costs they had quoted. We asked them whether the AMCincluded all charges which would be declared in the TER. Only seven

unds were able to answer the question on the phone. O these, all saidthe TER was not included in the AMC. Failure to declare the TER isparticularly perplexing given that rom June 2012, all unds covered bythe European UCITS Directive will be required to show their TER. Sincemost pension providers are also providers o UCITS products, it is disa-ppointing that they seem not to wish to give a ull cost breakdown whenselling pensions.

So, right now we would conclude that pension savers are not beingprovided with adequate in ormation. This means customers are ill-in ormed about the impact that costs have on the return on their pensionpots. This explains why our citizen juries were so shocked at pensioncharges. It is small wonder then that many people opt out o pensionsaltogether, while others end up with a pension pot that will not providean adequate retirement income.

Pension savers are not provided with clear in ormation about costs andwithout this in ormation, markets cannot work well.

Table 1: Declared charges rom selected providers

What chargeswould be leviedto the pensioninvestment?

What othercharges wouldbe leviedapart romthe AMC andadministrationcharges?

Are they ableto confrm thatno other urthercharges arelevied to theaccount?

Is itpossibleto get abreakdowno anycharges?

Pensionprovider 1

0.35 to 0.75dependant on und

None Yes only amcand admin charge

No

Pensionprovider 2



No

Pensionprovider 3

0.50 dependant onund

Possible In writing Pending

Pensionprovider 4

Dependant on und None Yes only amcand admin charge

No

Pensionprovider 5


No

Pensionprovider 6


No


13/30

11Levels o transparency

Pensionprovider 7

0.75 to 1%dependant on und

Possible In writing Pending

Pensionprovider 8

1 to 1.5%dependant on und


No

Pensionprovider 9

0.55% dependanton und


No

Pensionprovider 10

0.5% to 0.75dependant on und


No

Pensionprovider 11


No

Pensionprovider 12


No

Pensionprovider 13



No

Pensionprovider 14



No

Pensionprovider 15

0.10 to 06dependant on und


No

Pensionprovider 16



No

Pensionprovider 17


No

Pensionprovider 18

1% dependant onund


No

Pensionprovider 19



No

Pensionprovider 20

0.555 to 1%dependant on und


No

Pensionprovider 21 Approx 1.5dependant on und None Yes only amcand admin charge No

Pensionprovider 22

Approx 1.5%dependant on und


No

Pensionprovider 23

1.45% but may risein the uture


No

Administration and investment costs both matter, and i suppliers giveunder-estimated gures, surely this will only erode trust in the system.Even i many o the pensions sold are good ones, trust will break down

or the entire industry i there is a lack o transparency and providers are

unable or unwilling to disclose key gures.We do not want to single out individual pension providers, and indeed

we recognise there are many in the industry who have sought to addresssome o these issues. Un ortunately, collectively, the industry seems tomaintain a og around what goes on. Perhaps it eels this has been good

or business over the past 30 years so there is little incentive to change.However, 30 years ago de ned bene t pensions were the norm, and

costs did not matter to the pensioner because they had a promise on thelevel o pension they received, irrespective o charges. But in a worldwhere each individual saves or their own pension, costs are critical topension outcomes. It should be possible or market orces to allow saversto choose the best pension. But they need to know what they are buying.More transparency would empower the consumer and give them a airer

In a world whereeach individual saves or their own

pension, costs arecritical to pensionoutcomes


14/30


chance o understanding and navigating their way around the complexityo pension unds.

Small employer pensionsI retail pensions are sold unreliably, what about occupational pensions?It is a government policy, and one endorsed by the RSA and other stake-

holders, that employers are best positioned to choose pensions or theiremployees. In the past, o course, when the employer guaranteed the levelo pension paid, the employee had little concern about how much the pen-sion cost, since any short all would be made good. But in the new era o de ned contribution, any charges directly a ect the pension paid.

It is important that employers choose a good pension. In particular,with the introduction o auto-enrolment, many small employers willbe providing pensions or the rst time. So what do we know about thepensions which smaller employers are buying. Are they indeed providinggood value or money?

Very little evidence is available to answer this question. However,some relevant government research has been done in this area. In Table 2,we set out a review o various papers and a synopsis o comments and theresults. What does it tell us?

First, it only tells us about AMC, rather than the total costs o managinga pension. So employers, just like personal pension holders are without aclear understanding o the impact and the breakdown o cost charges.

The most comprehensive study o charges is entitled Charging Levelsand Structures in Money Purchase Pension Schemes .10 It had two parts,the rst o which involved talking to small trust-based employers . Theresearch ound charges to be quite high at 1.23% on average and 1.53% orvery small schemes. These gures did not include all commissions given tointermediaries, other elements o TER or any hidden costs. But the moststriking discovery in this study was the con usion surrounding pensioncharges. For example, nearly a quarter o scheme owners believed that thecharge was only levied on contributions, not on the outstanding bal-ance. O these, 45% thought the total charge to be only 1% or less on thecontribution; in other words, they believed it to be a tiny raction o thetrue cost. Some others believed that the charge was 5% per annum on theoutstanding balance, a gure which would leave little or the saver, while8% simply did not know what the charge was.

The second part o this study approached providers o contract-based

pensions. O 22 providers, only eight were willing to give gures ortheir charges. And the in ormation given was somewhat limited. O thegroup prepared to reply, they claimed 55% o the pensions they providedcharged less than 1%. The providers gave no in ormation o the distribu-tion o charges above 1%.

So what can we conclude? First, that charges are high. The numbersreported or small employers all exclude charges which would be includedin a TER, as well as hidden costs. Further, we eel respondents are likelyto have under-estimated, rather than over-estimated. Witness the numberwho believed there was no charge. But second, small employers, just

10. Charging levels and structures in money purchase pension schemes: Report of aquantitative survey , Department or Work and Pensions, 2010

It is a government policy, and oneendorsed by theRSA and other stakeholders, that

employers arebest positioned to6choose pensions

or their employees


15/30

13Levels o transparency

like retail purchasers, do not know what they are being charged. Theyo ten do not know the true AMC. They are there ore unlikely to under-stand the TER, still less the total cost. Essentially the employer, withgood intent, is trying to make provision or its work orce. But withouttransparency in charges and a clear understanding o the signi cance o those charges, there is a huge danger that it will be sold an expensive and

inappropriate product.Because o the danger o such misselling, one consultant hired by theDWP to research pension provision through small employers suggested itgives care ul consideration [to the view that] a light-touch, process-based approach to regulation in this area is not appropriate and moreheavy-handed, outcomes-based regulation would be better. 11

In act, the DWP has done the reverse. It has emerged that as part o the re orm to workplace pensions, the DWP has inadvertently allowedthe removal o the consumer protections which currently cover workplacepensions. It is now a matter o urgency that these are replaced; otherwise,in the absence o in ormation given openly to sophisticated buyers o pensions, there is a real danger o the otherwise admirable wider re ormsto pensions resulting in a misselling scandal.

Table 2: Recent studies on value or money o pension schemes

11. Default options in Workplace Personal Pensions: report of a qualitative study , DWPResearch Paper 628, Department or Work and Pensions, 2010. See table.

Title, year (andDWP researchpaper number)

Sample/approach Result Comment

Charging levelsand structures inmoney purchase

pension schemes ,2010 (630)

Small trust based DC schemes.

Quantitative.

Average charge level or trust-based schemes was 1.23%, orthose which had annual ee. Smallertrust-based schemes (with 611)members had an average chargeo 1.53% [page 28]. This coversintermediary commission in only50% o cases, so is underestimated.

But there was apparently asignifcant con usion on howcharges are levied. Almost a quartero respondents believed them onlyto be chargeable on contributions.Others believed there was nocharge, and some thought thecharge to be as high as 5%.

Like all studies, this only reveals theannual management charge, and notthe total cost o running the pensionplans.

However, this is the only quantitativesample o purchasers which isrelevant to auto-enrolment (viz smallschemes).

It reveals considerable con usionabout charges, and some very highcharges indeed.

Charging levelsand structures inmoney purchase


Suppliers o contract-basedschemes.

AMC charge or eight (out o22 approached) participatingproviders covering 3513schemes sold during the 2008/09fnancial year. All members; ieincluding active member discount.

No average charges provided; 55%charge less than 1% on averageor total sample. For schemes with

549 members, the average chargewas more than 1%.

This study could be relevant but itmust be treated with some cautionsince the in ormation came roma sel -selected group o providerswho would not have an interestin revealing high charges i theseexisted. The results were notaudited. Respondents did not revealthe distribution o charges over 1%.


16/30


De ault optionsin WorkplacePersonal Pensions ,2010 (628)

Literature review; 14 interviewswith providers (six), consultants( our) and other stakeholders.

N/A The report concludes with acautionary note that currentpractice may not be best practiceand that the DWP gives care ulconsideration [to the view that]a light-touch, process-basedapproach to regulation in this area isnot appropriate and more heavy-handed, outcomes-based regulationwould be better.

Current practicein workplace

personal pensions ;Pensions industry responses to theworkplace pensionre orms, 2009(591/592)

Qualitative study o providers (16)and intermediaries (24).

Questions about current contract-based schemes.

Typical AMC 0.9% includingintermediary ee. Range 0.41%but i member selected relativelyunusual und options, ee couldrise above 1% 0.5%, excludingintermediary ee.

AMC higher as a unction o scale,sta turnover, contribution size etc.

Notes the existence o activemember discounting, ie peoplecharged more when they leaveemployment.

Also quotes one respondent:Be ore stakeholder, there were ara t o charges some too high,some much too high, some OK.

A qualitative survey, with a sampleo providers and intermediaries(likely to be biased?).

Only looks at AMC.

Questions are about currentlarge schemes and, as interviewssuggest, unlikely to be relevant tosmall companies.

The use o vesting rules and de ault optionsin occupational


Qualitative study o 24 employers,10 providers and sevenintermediaries.

O employers, six had ewer than100 members, none had ewerthan 12.

Existing trust-based schemes.

Most charges between 0.4 and0.6%; range 0.150.9%

Employers said: The AMC waso relatively low importance incomparison to und per ormance inchoosing a supplier.

The sample chosen includesno pension plans o below12 members.

However, this is a very importantreport, since it appears to be theone which is relied upon by thegovernment in assessing ee levels.One might note that the aim o thestudy was not to discover current

ee levels, and that employers wereasked, in one hour, to answer over100 questions.

Costs o running pensions schemes:fndings o a

easibility study ,2007 (535)

Feasibility study, essentiallyqualitative o 35 DB andDC schemes o more than12 members.

Trust-based DC (fve responses)range 45133 bps.

Contract-based. Two employersdidnt know. Two said 0.81%.

Shows the e ort required to gleandata, since employers tend not tobe well in ormed about charges.


17/30

15Why dont customers demand change?

Why dont customersdemand change?

So ar, we have argued that retail customers and small employers are notwell-in ormed about costs, and are likely to be charged at a high level.This prompts the question, why have consumers not shown more inter-est? Surely, i they had been concerned then there would be mountingpressure by the industry to provide transparency? Further, what has beengovernments view o the cost-charging structures; why havent ministersintervened? These questions are now explored in the ollowing section.

Consumers and fnancial literacyAs we have seen, the e ect o costs on pension outcomes is substantial.But we have also seen that individual consumers lack awareness or under-standing o pension costs, hidden or otherwise. As a result, they do notgenerate the necessary pressure or change.

How should we account or this lack o consumer understanding.One possible explanation is consumers (lack o ) nancial literacy. Mightbetter nancial education solve the problem?

Financial literacy (o ten known as nancial capability in the literature)re ers to the the ability o individuals to make appropriate decisions inmanaging their personal nances. Research by Lusardi and Mitchell atthe Harvard Business School ound that consumers struggled with under-standing concepts such as stocks and bonds, compound interest and evensimple division calculations. 12

One reason pension charges are di cult to understand is because o the e ect o the compounding o charges. A question posed by Lusardiand Mitchell is particularly germane. They asked:

Lets say you have $200 in a savings account. The account earns 10%interest per year. How much would you have in the account at the end o two years?

Only 18% o respondents answered this question correctly. O thosewho answered it wrong, 43% made a simple interest calculation, ignoringboth principal and interest. Relating this to pension asset managementcosts, ailing to recognise that costs are compounded would mean people

12. Lusardi, Annamaria and Mitchell, Olivia S., Financial Literacy and Planning: Implica-tions for Retirement Wellbeing , Working Paper, Pension Research Council, WhartonSchool,University o Pennsylvania, 20 06


18/30

Seeing through the pension system16

were unable to understand the impact a small increase in costs wouldhave on their pension outcome.

Pension products tend to have a number o characteristics that makethem more complicated than other nancial products. They are by naturea long-term product and, since people tend to undervalue income in the

uture and overvalue income in the short term, a pension product may be

less attractive. This e ect is called hyperbolic discounting, meaning thatthe participants internal discount actor is too low or the uture and toohigh or the near uture rom the rational point o view. In other words,the loss incurred now (paying a premium) is overvalued, while the bene t(pension income a ter retirement) is under-valued. In addition, unlikeother products, there is no eedback mechanism when buying a pensionas the reward is only received a ter retirement, making it hard to learn

rom your mistakes until it is too late. Nor is it easy to compare pensionsrom the past generation. Comparisons are hard because schemes change

and the pension product is dependent on the per ormance o the nancialmarkets in a speci c time period.

Research by the Danish insurance association supports this evidenceo consumer behaviour. 13 It ound:

Most consumers do not know or care about their pensions Many consumers do not try to understand 90% o those questioned elt they were in good control o their

nances.

The Danish study went on to look at consumer motivation to engageactively in pensions. It concluded that any system needed to deal with twokey issues relating to consumer behaviour:

1. Level of choice. Questions arise as to how much choice consum-ers with low nancial literacy needed to have in complex mar-kets.Having many options attracts attention, but also reducedultimate action to purchase a pension.

2. The use of norms . People generally like to t in and do theright thing, so, unless pushed, consumers are unlikely to act or

eel motivated to act to change their pensions. As a result, theymaintain their current arrangements.

One long-term solution would be to educate consumers to under-stand their pension products. However, this also has a potential down-side. In 2006, the Financial Services Authority launched a seven-pointprogramme, the National Strategy or Financial Capability, to improvepeoples levels o nancial capability. In a review o this work, threebehavioural economists concluded:

I poor nancial capability is mainly a matter o psychology, the in or-mation-based approach o the National Strategy or Financial Capabil-ity is likely to have only a modest e ect in improving outcomes. The best

13. Pensions, nancial literacy and motivation, Danish Insurance Associationpresentation, 2010

Pension productsare by nature along-term product and, since peopletend to undervalueincome in the utureand overvalueincome in the short term, they may beless attractive


19/30

Why dont customers demand change? 17

empirical work nds that nancial education is not likely to have majore ects on knowledge and especially on behaviour. 14

So, nancial education is not a panacea. And whatever the level o nancial understanding, the market or individual pensions cannot be

e ective without transparency in clear and understandable English. Even

the most capable nancier cannot make a good choice i they do notknow what they are buying.

14. Financial capabil ity: a behavioural economics perspective , Consumer Research 69,Financial Services Authority, 20 08


20/30


What in ormationshould be provided?

In the previous sections we have learned that individuals and smallpension plans do not consistently know what they are charged or theirpensions, and/or they do not know the implications o those charges.As economists will tell us, where customers are unable to discover thecharacteristics o the product they are buying, it encourages a market orlemons, where poor products crowd out, and can eventually destroy, themarket or good ones. So, no one can object to pension buyers being givenin ormation explaining in clear and simple terms the costs associatedwith their pension. This in ormation should be:

Clear and easy to understand. Comprehensive. Capable of reconciliation. Like a bank statement, the customer

should be clear how the provider has arrived at the declarationsit has made.

Low-cost to produce. The in ormation should be provided sothat it does not place an administrative burden on the industrythat would be passed on in higher administration ees to the

und. So, ideally, it should use the same accounting systemsthat are already in place.

We note that in Denmark, such a system exists. In the ollowingdiscussion, we have borrowed rom Danish experience, and have outlinedits characteristics in more detail in Appendix 2.

We believe that customers need in ormation o two types. First, tohelp them make a decision about what pension to purchase. Second,

to monitor the activities o their pension provider during their pensionperiod, to check that it is being managed as expected.

The annual statementIn Appendix 1 we have illustrated a typical statement rom someone whohas a direct contribution (DC) pension. This statement should tell thepolicyholder about the charges which have been made to their pension. Itdoes not. Nor does it show the return be ore charges. Readers may wish torefect on this. A pension statement typically tells you (a) the amount savedat the end o the previous reporting period and new contributions made,and (b) the amount saved at the end o this reporting period. No in orma-tion is given about how the provider got rom (a) to (b).


21/30

19What in ormation should be provided?

We consider this practice to be wholly inadequate. Imagine a bankstatement that said: This is what you had last year, and this is whatyou have now, but we will give you no in ormation about how those twonumbers reconcile, in particular how much has been taken in ees. Thiswould clearly be unacceptable.

But it cannot be di cult or pension providers to show how the

numbers have been arrived at, because these must have been calculated inorder or the pension provider to know the level o pension saving at theend o the period. So, this in ormation can easily be made available to thesaver. I this was done online, the cost would be minimal.

That is what happens in Denmark. The saver can discover how theprovider has calculated the di erence between the money they had at thebeginning o the year and the contributions to it, and the money availableat the end o the period.

The Danish system is available online. It includes a summary scheduleo costs and, or each cost category, the customer can, at a click, nd abreakdown o each line item. Here is how we might envisage a statement

or a UK pension. We have also suggested some o the breakdowns whichwould be given or each line item.

Table 3: Per ormance evaluation table or UK pension

Money saved at the end o the last period 75,000

Contributions made [See Schedule 1] (this period) 5,000

Cash available or Investment 80,000

Returns Achieved [See Schedule 2] 5,000

Fund size be ore managers ees 85,000

Direct Costs o Investment (TER) [See Schedule 3] 1,000

Costs o other services provided [See Schedule 4]

Money saved at the end o the period (a ter ees) 84,000

Cost i you were to realise the investment today

Trans er Value o your pension 84,000

Schedule 1 , would allow the user to see when payments had been made into his or her account,and by whom.

Schedule 2 , would allow the saver to see how well the investment manager had per ormed andhow much had been spent by them on trading. We would envisage it to be shown:

Investment Returns

Total Gross Return rom Investment 6,000

Cost o trading etc. 1,100

Investment returns be ore managers ees(See evaluation below)

4,900

Other returns 100

TOTAL 5,000

It cannot be

difcult or pension providers to showhow the numbershave been arrived at, because thesemust have beencalculated in order

or the pension provider to know

the level o pensionsaving at the end o the period


22/30


Evaluation o Per ormance

Total Gross benchmark return 5,000

Value added rom investing be ore managers ees (100) (100)

Investment returns be ore ees 4,900

Schedule 3 , would break down the main elements o the TER, including management ees, audit,custodial ees etc.

Schedule 4 , would allow the inclusion o the costs o services which were bundled with thepension, or example li e insurance policies.

In addition to this per ormance evaluation table, customers wouldalso be given an indication o how much pension their savings might beexpected to provide them with. This is already given by pension provid-ers, assuming costs make no di erence to outcomes. Again, this shouldshow how the number has changed rom the previous year, and why thishas been the case.

In ormation be ore purchaseIn the section above, we have outlined the in ormation we believe shouldbe provided during the li e o the pension. But what should be providedto the customer be ore purchase? As we have noted, customers requently

ail to understand the implications o costs on pension outcomes. We sug-gest that these be made more explicit.

As automatic enrolment approaches, employers are in the ront line inchoosing pension schemes or their employees. We agree with NAPF thatit is vital that employers receive clear and comparable in ormation aboutcharges be ore they choose a scheme or their employees. We welcomethe approach to this set out in the recent consultation document rom theNAPF-led working group. 15 In parallel, we have been considering whatin ormation should be given to individuals when they are consideringjoining a scheme.

In a typical sale o a pension, a customer will be given in ormationabout what level o saving is required to achieve a given pension. So, orexample, they will be told that, on given assumptions about returns,a certain level o saving will give a certain level o pension. Beside this,a projection should be given or the value o the pension if no fees werecharged , ie i the TER was zero.

We would note that this is a generous assumption or active undmanagement, since it assumes that the trading o securities is value addingwhen all participants have been included. Most academics would suggestthat costs o trading might be added to TER to illustrate the likely di er-ence in pension outcome.

Such in ormation should be mandatory in any sale.

What other in ormation do consumers need?We have suggested there are two speci c pieces o in ormation whichpension savers should receive.

15. www.nap .co.uk/PolicyandResearch/DocumentLibrary/~/media/Policy/Docu-ments/0254_Telling_people_about_DC_pension_charges.ashx


23/30

21What in ormation should be provided?

1. An annual statement which reconciles their investment savings.2. A statement be ore purchase which shows the likely e ect that

ees will have on their pension outcome.

Neither o these seem radical demands. Indeed, without such in orm-ation, clearly expressed, it is di cult to see how anyone can understand

whether their pension is appropriately managed, and represents valueor money.To these requests we would add a urther one. The pensions industry

in the UK, like that in Denmark, should aim to provide one plat ormwhere all the pensions which an individual holds are reported upon in thismanner. People hold many pensions rom di erent organisations, andshould have the ability to compare and contrast them. In Denmark, sucha system is already in operation. In Britain, where we boast one o themost advanced und-management industries in the world, it should surelybe possible to replicate what is already available on the other side o theNorth Sea.


24/30


Summary andrecommendations

For markets to work e ectively, consumers need to know what they arebuying. In the purchase o pensions, it is clear many do not. Indeed wehave shown that individual savers are not in ormed about costs, and thatsmaller occupational unds also lack an understanding o charges andtheir potentially pro ound e ect on pension outcomes.

So, i we are to avoid a market or lemons, where good suppliersare crowded out and people stop saving, simple, accurate, high qualityin ormation needs to be provided.

There are some short-term xes which should to be put in place. First,all pensions suppliers must commit, rom today, that their sales orces willgive ull in ormation on total expenses and their implications or pensionoutcomes. Our market research into the lack o in ormation given tocustomers suggests there is a huge gap between the policies advocated inthe board room and what is happening on the ground. This gap needs tobe lled immediately.

This is particularly true or occupational pensions, where the ben-e ciary does not choose the pension provider. In our view, expressedin previous RSA reports, and endorsed by the Parliamentary Workand Pensions Committee, limits like those applied to stakeholderpensions need to be placed on the charges which can be taken romoccupational pension savers. Without such protection, the UK risksa misselling scandal.

Other measures go beyond the short term. For example, we haveshown that much o the in ormation given to pension customers is pro-vided in a ormat that is hard to understand.

We need a change to the way annual statements and pre-purchase

in ormation are given to customers. The annual statement should bepresented like a bank statement. This is simple to do. It already happensin Denmark, and we have shown what the statement might look like.

Pre-purchase in ormation should show what e ect charges will haveon pension outcomes.

Both o these seem the most basic requests. And, i a national systemo this type was established, the market or pensions saving would beginto work better.

These are all simple recommendations. That such in ormation doesnot already exist must be a contributory actor to the lack o con dencepeople have in the pension system. These are simple, low-cost, nudgesuggestions which will bene t all pension suppliers wanting to succeedon the basis o the value o the products they sell. But, more important,

I we are to avoid a market or lemons, where

good suppliers arecrowded out and

people stop saving,simple, accurate,high quality in ormation needsto be provided


25/30

23Summary and recommendations

by providing this in ormation, Britons can have greater con dence in thesaving system and enjoy better, higher pensions at reduced cost.


26/30


Appendices

Appendix 1: An example taken rom a typical

UK pension statementThese are sent annually by most providers to de ned contributionpension holders in the UK. We have put this together rom a real clientstatement to illustrate the simple di erences between the in ormationavailable in the UK and in Denmark ( gures and calculations areillustrative). We note that the UK in ormation is summarised rom oneo the countrys most reputable suppliers, and has been taken rom a six-page statement. No summary table, such as the one compiled is includedin the statement.

Yearly statement

Prepared or Mr Hansen

This pension plan aims to build up money in a tax-e fcient way or when you retire. It will thenprovide a pension or the rest o your li e.

The statement shows what your plan is worth now, the payments into your plan and how muchpension you might get. It gives you the opportunity to review your fnancial needs.

Please keep this statement sa e or your uture re erence.

Your plan Summary ( rom 1 April 2011 to 1 April 2012)

Plan Values Calculated on 1 April 2012

Current Value 30,000

Last years value 20,000

The value is the amount you could have trans erred to another plan

Plan Payments or 1 April 2011 to 1 April 2012

Payments into your plan 9,000

What you might get back

I our investments grow at: 7%

This could give you a fnal plan value o 50,000

This could give you a taxable pension each year o : 2,000

Detail given o assumptions or the calculation

Investment gain or the year 1,500

Charges (broken down by month, but not by cost category) 0.5%


27/30

25Appendices

Appendix 2: The Danish systemIn Denmark, consumers are able to log on to a government-backed sitethat provides them with in ormation o their pension provision (www.pensionsin o.dk). The site, which is now subscribed to by most pensionsuppliers, provides comprehensive in ormation to all Danish citizens. Itexplains how the pension system works, and once logged in, it provides

individuals with details o all their pensions. For example, the site pro-vides in ormation such as:

How the consumer is covered, in which unds they are invested. What contributions have been made into the und. What returns have been made. What has been paid out in costs and expenses, or each element

o the pension ( or example, or li e insurance). How much money they currently hold in the und and

its present value. And, as with a UK statement, it illustrates what the pension

might be worth in the uture.

An example o how this might look or the individual consumer isgiven below.

The Danish Pension statement

Figure 1: Example o pension statement

Eksempel: Hr. Hansens pensionoversigt

Ind Ud Total

Betaling for administration

Betaling for forsikringsdaekning

40.000 kr.

11.000 kr.

2.000 kr.

8.000 kr.

271.000 kr.

230.000 kr.Depot pr. 1/1

Indbetaling

Rente

Depot pr. 31/12

Translation

KrBalance at last statement 230,000

Deposits 40,000

Returns 11,000

Costs or administration 2,000

Costs or insurance coverage 8,000

Balance at year end 271,000

For each o these categories, the account holder can then obtainurther in ormation and detail on the unds. This depends on the type

o investments that the port olio has. In terms o transparency, the de ni-tions given or costs and the gures that have to be included areas ollows:


28/30


Direct and indirect administrative costs. These can includeund trustee and administration expenses, along with general

set costs. Direct and indirect investment costs. These include the actual

trades made on the customers account, such as any money thecompany receives rom investment managers, per ormance ees

charged by und managers and ees on stock lending. In real-estate unds, these might also include other expenditures relatedto the properties current income, such as expenditure or col-lecting rent, salaries o caretakers, property taxes, maintenance,improvements and any other administration costs.

Annuity costs. The costs associated with buying an annuityin a pension.

Risk-free return. The risk o return is calculated as the excessreturn over the risk- ree rate the capital could obtain on marketconditions.

The system provides the account holder with the same level o detailand in ormation that the pension und itsel has on its costs, its per or-mance and its benchmarks. By visiting the website, you will be able to seesome demonstratory explanations o the system, highlighting its ease o use and also its level o detail.


29/30


30/30

seeing through the british pension system

Documents