The AMBACHTSHEER letter

TOWARDS ‘INDIVIDUALIZED SOLIDARITY’ IN PENSION DESIGN:WHY THE DUTCH CAN AND SHOULD LEAD THE WAY

“Integrative thinking is the ability to hold two opposing ideas in mind at once, andthen to reach a synthesis that contains elements of both but improves on each.”
Roger Martin

Pension Turmoil in the Netherlands

The Dutch have been thought-leaders in thedesign and management of collective pensionsystems for a long time. Their ‘collectivity’origins reach way back to the ‘polder model’,which reflects a special sort of Dutch solidarityforged by working together for centuries to keepthe sea at bay. After the Dot.Com bubble morethan a decade ago, pension regulator DirkWitteveen declared in 2002 that, without majorreforms, the Dutch pension system could end upunder water too. Despite strong protests at thetime, stronger funding rules were instituted, andit seemed that the system was healthy onceagain …. until the Global Financial Crisis struckin 2008/9.

The GFC raised even more fundamental questionsabout the 21st Century sustainability of theDutch pension system (and for that matter, ofevery other pension system on earth too!). As aresult of these questions in the Netherlands, thesearch for more sturdy models that could meetthe ‘21st Century Sustainability’ test was on. Inour view, the Dutch have strong comparativeadvantages to be a global leader in pension innovation:high public interest level, strong pensionexpertise, strong pension institutions, and thatcenturies-old collective approach to problemsolving.

As we write this, two new pension model candidateshave surfaced over the course of the lastfew years, called ‘the nominal contract’ and ‘thereal contract’ for short. Both models have theirchampions and detractors. But there is also agrowing camp arguing that neither model scoreswell against such sustainability criteria as affordability,payment certainty, fairness, clarity ofproperty rights, sensitivity to individual preferences,and pension contract understandability byplan participants. We are in the ‘neither model’camp, and the brave goal of this Letter is to arguefor an alternative by taking a number of apparentlyopposing ideas (e.g., enforcing solidarity vs.accommodating individual preferences at thesame time) and to, in Roger Martin’s words,”reach a synthesis that contains elements of bothbut improves on each”.

Why ‘Individualized Solidarity’ is Not aContradiction

In his book “The Opposable Mind” (HarvardBusiness School Press, 2007) Martin assertspeople tend to address most challenging problemswith ‘either-or’ mindsets …. when in fact ‘andand’solutions are often superior to forcing adecision between Choice A and B. In our view,so it is with ‘individualization’ and ‘solidarity’ inpension system design. The ideal pension designencompasses both elements. The challenge is tothink through which elements of the systemshould accommodate individual preferences, andwhich elements are better addressed collectively.

A good ‘individualize’ example is investmentrisk. This kind of risk means very different thingsto the typical young worker (“how much can Iafford to save now for decent retirement 40 yearsfrom now?”) and the typical pensioner (“Will Iget my pension next month without any reduction?”).These two questions graphically illustrate that any pension formula that assumes uniform investment risk tolerance among all plan participants will leave everyone dissatisfied. The right application of ‘solidarity’ in this case is to offer plan participants separate long-horizon returnseeking, and shorter horizon payment-certainty investment tools, and to systematically transition exposure from the former to the latter over time.

A good ‘collectivize’ example is longevity risk within age cohorts. Within any group of same-age pensioners, for example, some will have short lives, and others long ones. But who will be in the short and the long ends of the mortality distribution is not known ahead of time. Pooling this uncertainty means annuities can be priced based on average life expectancy to the benefit of all. Another good ‘collective’ example is for large groups of pension plan participants to use the same pension management organization. Large organizations can manage investments and deliver pension administration with higher skill levels and greater economies of scale for the benefit of all.

Foundations for an ‘Individualized Solidarity’Pension Model

So what does a pension model that scores high onsuch sustainability criteria as affordability, paymentcertainty, fairness, clarity of property rights, sensitivityto individual preferences, and pension contractunderstandability by plan participants looklike? We start by drawing on the wisdom of the ages,personified by the genius of Albert Einstein(relativity theory), John Nash (game theory), Jan
Tinbergen (public policy theory), John MaynardKeynes (public policy theory), Peter Drucke(governance theory), and TIAA-CREF (pensionmodel design and implementation):

1. Albert Einstein once remarked: “Makethings as simple as possible, but no simpler”.In our view, most Dutch pension‘contracts’ today cannot pass the Einsteintest. They are too complicated for nonexpertsto understand. Worse, the currentreform proposals create the risk these contractswill become even more complicated.This will reduce the already-declining publicconfidence and trust in the Dutch pensionsystem even further.

2. John Nash warned: “Beware of bargainingarrangements that have potential ‘win-lose’outcomes embedded in them … they willeventually become adversarial”. Most Dutchpension arrangements today do not haveclear property rights (e.g., the size andcertainty of future balance sheet claims ofyounger and older plan members at anypoint in time are typically not fully defined).Nash’s game theory model predicts thatwhen adverse economic conditions such as2008/9 arise, competing positions about theownership of balance sheet assets and liabilitieswill arise. This has in fact come to pass.Looking ahead, the new ‘real contract’ proposesto perpetuate this ‘win-lose’ problemby using a discount rate curve based onthree non-market-based, subjective parametersto determine how money is divided betweenyounger and older plan beneficiaries.

3. Jan Tinbergen proved: “The number of economicgoals to be attained must be matchedby the number of instruments capable ofachieving them”. We noted above that twofundamental economic goals of pension systemsare 1. Affordability, and 2. Paymentcertainty.The Tinbergen rule states thatachieving these two goals will require twofinancial instruments (i.e., a longer-term,wealth-creating instrument for affordability,and a shorter-term liability-hedging instrumentfor payment certainty). Let’s call pensionmodels that meet this dual test ‘2 goals-> 2 instruments’ models.

4. John Maynard Keynes observed:”Institutional investors seem more interestedin winning adversarial trading games(‘beauty contests’) amongst themselves thanin creating long term wealth for their clients”.Too many pension organizations continueto engage in zero-sum, adversarial,‘beauty contest’ investment games, and toofew are engaged in longer term wealthcreationstrategies. The ‘2 goals -> 2 instruments’pension model offers a clear, unambiguousrationale for adopting explicit longerterm wealth-creating investment programs.While such programs may be ‘risky’in a short horizon context, they are muchless so for multi-decade holding periods. Atthe same time, the payment-certainty instrumentmust give older workers and pensionerscomfort that the deferred annuity contractsthey have purchased will indeed paythe contracted amount.

5. Peter Drucker wrote: “Pension organizationsneed effective governance disciplines just asmuch as any other organization”. A growingbody of research is confirming this reality.Only pension organizations with effectiveboards and managements can serve plan participantsas well as they have a right to expect.

6. TIAA-CREF demonstrated: that large pensionsystems can successfully manage a ‘2goals -> 2 instruments’ pension model for avery long time (e.g., since 1952). CREF permitsparticipants to build a pension pot overlong holding periods. TIAA permits participantsto buy payment certainty through deferredannuities. A 2009 study showed thatin the sample of 77,000 active plan participants,all age-cohorts were on track toreplace at least 70% of their preretirementincome (including the Social Securitypension). See Hammond and Richardson,”Staying on the Path to a Secure Retirement”,TIAA-CREF Research Institute.Peter Drucker, a TIAA-CREF participant formany years, wrote approvingly about its ‘2goals -> 2 instruments’ model in his 1976book “The Unseen Revolution”.

Key Features of the ‘Individualized Solidarity’Pension Model

All this translates into the reality that sustainable21st Century pension arrangements have threekey design features:

1. A Long Horizon Return-Seeking InvestmentInstrument: in the spirit of Keynes’ investmentvision, such an investment programseeks, acquires, and nurtures sustainable,growing long horizon cash-flows in the formof dividends, rents, tolls from a diversifiedportfolio of public and private investmentvehicles. The fact that ‘the market’ will valuethese cash-flows differently from day today is irrelevant. Eventually (e.g., for 10-20year+ holding periods), as long as the aggregatecash-flow of the long horizon portfolio
performs as expected (e.g., grows in excessof the rate of inflation), ‘the market’ willvalue the portfolio on its economic merits.These programs are managed by engagedinvestors who, in the spirit of the HeisenbergPrinciple, positively impact investment outcomesthrough their individual and collectiveengagement strategies with investeeorganizations (e.g., public or private corporations).For more on this style of investing,see our September-December Letters lastyear, and the January-July Letters of thisyear.

2. A Liability-Driven Payment-Certainty Instrument:it supplies life-long payment certaintyin the form of life annuities, whichplan participants purchase at a ‘fair-value’price (i.e., reflecting the actual structure ofinterest rates at the time of purchase andconservative longevity expectations). Thebalance sheet of this payment-certaintyinstrument is managed and regulated to ensurethat payment promises made will bepayment promises kept. In the spirit ofEinstein’s ‘keep things as simple as possible’dictum, there is only one simple (i.e.,explainable), market-hedge-able form ofannuity on offer. Members begin to purchasethese annuities mid-career on a deferred basisand accumulate them gradually over a,say, 20-year period. There is nothing new inthis, as current pension contracts are alreadydesigned to do this. What is new is thatyounger members no longer overpay, andolder members no longer underpay for theirdeferred annuity purchases. Again, see priorLetters for more on this style of liabilitydriveninvesting.

3. A Life-Cycle Transition Protocol: it startsfrom the reality that people journey throughthree life phases: pre-work, work, and postwork.A target post-work standard of livingis financed in part by a national pay-go oldage pension component, and in part byworker savings and the investment return onthose savings. Plan members receive regularupdates of progress towards achieving thetarget pension on the target date. A defaultrule determines members’ allocationsbetween the two investment instrumentsover the course of the work and post-workphases of their life-cycle (e.g., in the exampleabove, the purchase of deferred annuitiesstarted mid-career and continues for a 20-year period). See last year’s August Letterfor more on designing a functional life-cycletransition protocol. Also, see our policypaper “The Canada Supplementary Pension Plan” (CD Howe Institute, 2008) on creating a collective pension arrangement for all Canadians without a work-place pension plan for even more detail. Some Dutch pension experts have been advocating the ‘combi-contract’, which has some of the key design features of the ‘2 goals -> 2 instruments’ model, but still suffers from a lack of property rights clarity.

Of course, just agreeing to move to a ‘2 goals -> 2instruments’ pension arrangement doesn’t get youthere. It also requires a well-thought out, wellexecutedtransition plan to get from here to there.

Getting from Here to There

Three steps must be agreed on to transition currentcollective Dutch pension arrangements to ones thatmeet the sustainability criteria of affordability,payment certainty, fairness, clarity of propertyrights, sensitivity to individual preferences, andpension ‘contract’ understandability by plan participants:

1. Create a protocol to convert current accruedcollective pension rights of plan participantsinto ‘2 goals -> 2 instruments’pension rights: this protocol needs to passthe dual tests of understandability, and bothactual and perceived fairness to all participants.Devising such a protocol will behard, exacting work; but it only has to be done once.

2. Re-write pension laws to ensure ‘2 goals -> 2 instruments’ models are legal: this willrequire legal expertise, common sense, andcommon ‘greater good’ purpose.

3. Engage pension plan participants in theprocess: the transition to ‘2 goals -> 2 instruments’pension models will not happenwithout broad public support. Gaining thissupport will require a radical communicationstrategy rethink. The Dutch media reportingof the pension reform debate thusfar has been unhelpfully convoluted. Asimportantly, some of the pension experts involved in the reform debate appear to have been more interested in displaying their technical virtuosity than in using plain language understandable by the public at large.

A footnote to these three transition steps is the ongoingneed to continue to raise the governancequality of Dutch pension organizations at the sametime. Moving to the ‘2 goals -> 2 instruments’model does not impact organizational scale in anyway. It does, however, help clarify the specialistskill sets needed for organizational success.

Why Should the Dutch Lead?

Through his 2002 declaration, Dirk Witteveen(deceased in 2007) was the first public officialanywhere in the world to sound the alarm that theglobal pension environment had changed, and thatpension arrangements would have to adapt to it.Since then, these adaptation processes everywherehave been unfolding painfully slowly. Throughtheir culture and their expertise, the Dutch havestrong comparative advantages to be the firstcountry in the world to successfully adapt theirpension system to the longer term global realitiesof aging populations, rising longevity, slower economicgrowth, and lower investment returns. Anew Dirk Witteveen must step forward.

One final point. Some argue that the Dutchpension reform window is closing, and that it isbetter to implement the proposed (but flawed)‘nominal’ and ‘real’ contracts now, rather thanspend more time finalizing a contract that can trulypass the sustainability tests of affordability, paymentcertainty, fairness, clarity of property rights,sensitivity to individual preferences, and pensioncontract understandability by plan participants. Wefavour the opposite argument. It is far better tospend a little more time getting the contract rightnow, than spending much more time later cleaningup an even bigger pension mess. Or, to quote aDutch friend: “Beter ten halve gekeerd, dan tenhele gedwaald!”

CARPE DIEM!

Wilt u meer weten over pensioenen?
Tijdens het Pensioen Forum op 1 en 2 oktober 2013, komen bestuurders en pensioenexperts bijeen. Dit is dé ontmoetingsplaats voor beslissers van pensioenfondsen, pensioenuitvoerders en verzekeraars. Meer informatie over het programma en de meer dan 55 sprekers vindt u op de website van Het Pensioen Forum.