SUPLEM. INGLES:SUPLEM. INGLES, page 1 @ Preflight€¦ · Although a progressive change in the...

Between two decades

Every day, every year seems to speed by more quickly and the time has now come, at the end of the first decadeof the third millennium, to sit back and take stock.

The new millennium loomed up under a pall of dread and doom. The Y2K bug was going to wipe out ourcomputer-based society overnight. This spurious threat now seems risible in comparison to real catastrophes that didoccur in this first decade, like 9/11 in the US or 11-M in Spain, the swine-flu panic, the devastating aftermath ofhurricane Katrina and the more recent earthquakes of Haiti and Chile or the disastrous oil spill caused by theexplosion of the Deepwater Horizon oil rig. The financial collapse and the consequent swathe of failing businesseshave by now become a byword for our times. The recent history of this 3652-day period therefore offers up theparadox of a world that is becoming more socially and financially shaky as it becomes more globalised.

Our misplaced trust in an ostensibly fault-proof system has been one of the hardest shocks to our system inrecent history. It put paid to the long twentieth-century dream of breakneck consumerism and vanity. The sackclothand ashes years of this early twenty-first century have brought us crashing back down to earth. But necessity is alwaysthe mother of invention and we now need to rethink our strategy so far, trying to bring home the value of riskmanagement to all organisations.

Although a progressive change in the «risk appetite» of business organisations has been appreciable for sometime, the results of the 5th risk benchmarking survey of the European Federation of Risk Management (FERMA) –presented in the last four months of 2010 – shows that clients are in general reasonably satisfied with currentinsurance prices and trust in their insurers to help them develop their business and expand worldwide. It also reflectsthe respondents’ desire for more support from insurers in managing new and emerging risks, especially for exposurescurrently considered to be «uninsurable».

Our interviewee in this issue is Belén Kortabarria, Financial Manager of Mondragon, the biggest business groupin the Basque Country and the seventh biggest in the whole of Spain. She takes up this theme, seeing no restrictionwhatsoever in the insurance risk-transfer capacity, not even for taking on the risks of the over 100 co-operatives andover 200 productive subsidiaries abroad, with the caveat that «perhaps the most worrying risks are those we don’tknow about yet».

The first of the studies in this issue takes an academic slant on the future reform of road-accident compensationarrangements, mooting lower compensations for minor injuries in favour of “financing” substantial compensationincreases for major injuries and hence better protection for road-accident victims overall.

The second study looks at the application of credibility theory, through two different models, to the calculationof a health insurance premium, trying to tally the premium with the known claims ratio of the insurance productconcerned.

The third and last article reflects the thoughts of François Settembrino, honorary president of FERMA andmember of this review’s editorial board. In three words with identical initials –risk, respect and responsibility– he sumsup his ideas for rethinking Risk Management.

The report drawn up by FUNDACIÓN MAPFRE presents the ranking by premium volumes of LatinAmerica’s 25 biggest insurance groups in 2009. This eighth ranking survey breaks down the figures into life, non-lifeand total.

The Claims Observatory, based on the earthquake of magnitude 8.8 in the Richter scale that hit Chile inFebruary this year, presents a solution for guaranteeing continuity of the business activity after catastrophes of thatscale, using Business Continuity Planning.

So the year is running to its end and, as usual, there will be some things we never got round to doing. Maybethis will stimulate us to take them on with renewed vim next decade.

Merry Christmas and a happy new year

iesgosR

SE

PT

EM

BE

R/D

EC

EM

BE

R 2

010

ENGLISH APPENDIX

YE

AR

XX

VII

G e r e n c i a d e

y Seguros 108

GERENCIA DE RIESGOS Y SEGUROS • Nº 108—20102

➜Directive 2005/14/EC of the EuropeanParliament and of the Council of 11th.May, 2005, relating to insurance againstcivil liability in respect of the use of motor vehicles(Fifth Directive), modifies certain aspects of thefour previous directives with the clear purpose ofreinforcing the protection of those injured in roadtraffic accidents, as well accelerating the payment ofindemnities. In order to incorporate theaforementioned directive into our legislation, Law21/2007 of 11th.July was approved and re-adaptsthe text of the Law on Legal Liability and Insurancefor the Use of Motor Vehicles (Ley sobreResponsabilidad Civil y Seguro en la Circulación de

LLUÍS BERMÚDEZ MORATAMERCEDES AYUSO GUTIÉRREZ MIGUEL SANTOLINO PRIETOUniversitat de Barcelona

REFORM OF THE SCALE

OF INDEMNITIES

REFLECTIONSand outlook on the future

GERENCIA DE RIESGOS Y SEGUROS • Nº 108—2010 3

LAT

INST

OC

K


although the new Law has increased the limit of thecoverage for bodily injury up to 70 million Eurosper claim, in most cases, there will not be a greaterguaranteed protection for those injured in trafficaccidents if the binding system of evaluation thatdetermines the indemnity limits for these claims ismodified. In other words, a reform of the evaluationsystem for personal damages caused by trafficaccidents is imposed. In this regard, at thebeginning of 2008, the Director of Insurances andPension Funds considered that the updating of thescale needed maximum consensus, calm, debate andreflection, and studies on the impact that the reformcan have on the insurance sector.

In response to the request of the Director ofInsurances and Pension Funds and from the neutralposition held by university professors, we decidedto analyze the ins and outs of the evaluation ofpersonal damages caused by traffic accidents and, inthis way, provide a tool to support the necessarydebate and reflections on this topic. In this task, wereceived the cooperation of the MAPFREFoundation, via the Insurance Research Grants(2007 year), which led to the publication of the

Vehículos a Motor - hereafter LRCSCVM), approvedby Royal Decree 8/2004, of 29th.October.Following the recommendations of the FifthDirective, and in order to guarantee the effectiveprotection of road traffic accident victims, the newlaw has raised the coverage limits of the obligatoryinsurance up to 70 million Euros per claim for ofbodily injury, regardless of the number of victims.With regard to improving the processes of settingthe amount of the indemnities, the new Law hasincorporated two new concepts into our legalregulations: the reasoned offer / reply procedure.(Xiol, 2009)

The LRCSCVM establishes the obligatoryelement of motor insurance and defines it as legalliability insurance for damages arising out of trafficaccidents. The insurer is obliged to indemnify thirdparties for damages caused by the insured whilstdriving a vehicle. As a result, insurers undertake animportant social role since they are responsible forcompensating the victim financially for personalloss which is interpreted as loss of quality of life andany suffering created. Nevertheless, to determinewhat amount is a «fair» compensation that theinsurer must pay to the victim of a traffic accident,as an indemnity for the bodily injury suffered, isnot a simple task nor exempt from discussionamongst the experts.

In Spain, from 1995, the evaluation of theindemnity for damages suffered by the victims oftraffic accidents had to be quantified in accordancewith the «System for the Evaluation of Damagescaused to Persons in Traffic Accidents» published inthe Annex of the LRCSCVM and commonlyknown as the scale of indemnity. This evaluationsystem is articulated by means of tables which showthe concepts and amounts to be awarded.

In view of the change in rules derived fromthe Fifth Directive, it would seem logical to reflecton the determination of the indemnity awardsunder the application of the Spanish scale. In effect,


➜THE UPDATING OF THE SCALE NEEDED MAXIMUM CONSENSUS, CALM, DEBATE AND REFLECTION, ASWELL AS STUDIES ON THE IMPACT THAT THE REFORM CAN HAVE ON THE INSURANCE SECTOR

evaluation of its impact after defining two simplereform approaches. Finally, in the conclusions, wesummarize our opinion on the reform of theevaluation system of personal injury derived fromroad traffic accidents.

CURRENT SCALE: MOTIVESAND OPERATION

On the above subject, we believe that, beforeinitiating the debate on the reform of theindemnity scale, it is necessary to bear very much inmind the situation and motives that led thelegislator to include a «System for the Evaluation ofDamages caused to Persons in Traffic Accidents» inthe Annex of Law 30/1995 on Legal Liability andInsurance for Motor Vehicles and Road Traffic. Inthis way we can consider the validity of theassumptions that inspired the current scale whichwas implemented to respond to a series of factorsthat arose and that we summarize in the followingparagraph.

work «Perspectives and Economic Analysis of theFuture Reform of the Spanish System of Evaluationof Bodily Injury» (Bermúdez, Ayuso and Santolino,2009). The content of this publication wasstructured in two parts. In the first part, wepresented the theoretical concepts that help tounderstand the dynamics of how the scale works inSpain, comparing it with systems in other countriesaround us, as well as to get as clear as possible anidea as to the position of the system reform. In thesecond part of the work, we decided to respond tothe need to carry out practical exercises ofevaluating the economic impact in the context ofdifferent scenarios that the future reform of thescale might define.

This article, based on the work mentioned inthe previous paragraph, is intended to summariseour conclusions and reflections on the way inwhich the scale works and its possible reform. Inthe following paragraph, we recall the principlesand objectives that inspired the current scale andexplain how it works. In the third section, weanalyze the influence that the Europeanharmonising framework could have on the reformof the scale and, after, in section 4, we reflect on thereform of the scale perspectives, bearing in mind itsmultidisciplinary character (legal, medical andsocio-economic). Having reached the conclusionthat, in order for the reform of the scale to provideguarantees, it is necessary to take into account theeconomic impact that the different reformproposals have on the overall claims settlement cost,in the fifth section, we carry out an economic


According to UNESPA data, for the 1991 year,the combined underwriting ratio for motorinsurance was 108 per cent on earned premiums.One of the main causes for this result was thecontinuing tendency by judges, from the mid 80´s,to increase the indemnities for personal damages(judicial inflation). This phenomenon, together withthe differing criteria applied by different judges,discouraged amicable settlements between theparties. The excessive litigation led to delays in thecourts in the fixing of indemnities for claims withbodily injuries, accounting for a significantdifference between the amount of the indemnitypaid and the amount that had been used to calculatepremiums some years earlier. In other words, thereserves of insurers, calculated from the premiumsearned in the year of occurrence of the accident,were insufficient to pay for the final indemnitysettlements paid several years after continuous legalproceedings. Insurers were not the only partiesaffected by this situation: the courts of justicecollapsed; traffic accident victims were compensatedafter long delays and with varying levels ofindemnity, depending on the court assigned; andthe regulator experienced serious difficultiescontrolling the solvency of the companies thatspecialised in this class of insurance.

With the approval of the scale, the legislatorintended to end the situation described in theprevious paragraph by providing all the partiesinvolved in the process with mechanisms ofpredictability, moderation and equality in theevaluation of the monetary indemnity for damages

suffered by victims of traffic accidents. Theindemnity scale is to be used for the evaluation ofbodily injury (death, permanent and temporaryinjuries) suffered by the victim of a traffic accident.The evaluation system operates via a system oftables which include all concepts susceptible toindemnity together with the rules or criteria for themonetary evaluation of the aforementionedconcepts - all this, within a contained scheme, i.e.without the possibility of operating outside thesystem. In order to achieve the declared goals, thelegislator imposed the obligatory nature of itsapplication by judges. Since its approval, thebinding nature of the scale of indemnities has beenthe most argued and controversial aspect.Nevertheless, following the Constitutional Court

➜WITH THE APPROVAL OF THE SCALE, THE LEGISLATOR TRIED TO PROVIDE ALL THE PARTIES INVOLVED IN THEPROCESS WITH MECHANISMS OF PREDICTABILITY, MODERATION AND EQUALITY IN THE EVALUATIONOF THE MONETARY INDEMNITY FOR DAMAGES SUFFERED BY VICTIMS OF TRAFFIC ACCIDENTS


party. For all contingencies, the design of the tablesessentially contemplates two types of personaldamages to be compensated for road trafficaccidents: financial and non-financial damages.

The basic indemnity in case of death (Table Iof the scale) is calculated according to the numberof injured persons and how they are related to thevictim, on the one hand, and the age of the victimon the other. The corrective factors for the basicindemnities by death appear in Table II andcontemplate percentages increases on the basicindemnity depending on the financial damages thatoccurred (the net annual income that the victimwas earning from personal work), as well as anyspecial family circumstances.

The basic indemnities for permanent injuriesare shown in Table III. In this case, the basicindemnity is calculated according to a points systemreflecting the severity of the injury. The score startswith Table VI (medical scale), differing between thepsychophysical prospects (after-effect) and theaesthetic damage. Finally, the calculation of thequantity to be indemnified will depend on themonetary value of the applicable point in the tableand the age of the victim. The corrective factors forthe basic indemnities for permanent injuries areshown in Table IV. In the same way as for death, thefinancial consequences suffered by the victim aretaken into account. In these cases, there is also thepossibility of including additional compensation forpain and suffering and for the degree of normaloccupational incapacity caused by the accident, aswell as other corrections related to expensesincurred as a result of the victim´s disability.

The indemnities for temporary disabilityappear in Table V. The basic indemnity is a fixed anddaily amount according to the days off work(distinguishing between hospitalization,preventative and non- preventative days off-work).The correction factors related to financial loss alsoappear in the same Table V,

judgement 181/2000 (STC 181/2000) this pointhas stopped being debated, at least openly.

The System for the Evaluation of Damagescaused to Persons involved in Road Trafficaccidents is structured in two sections plus an«appendix» (appendix to Annex of the Law). In thefirst section, the criteria for determining legalliability and indemnity are considered and, in thesecond one, the rules on the application of thesystem. After defining these criteria, principles andrules, the table of the scale is presented in appendixand is divided according to the type ofcontingency: 1) Indemnities for death (Tables I andII); 2) Indemnities for permanent injury (Tables III,IV and VI); and 3) Indemnities for temporarydisability (Table V).

In summary, the evaluation method begins byfixing basic indemnities for death, permanentinjuries, or for temporary injuries, and then appliescorrective factors according to the concurrence ofcertain circumstances, basically related to thefinancial and family circumstances of the injured


So, the evaluation and repair of the bodilyinjuries in Europe are based on very different legaldispositions, and on very different jurisprudentialand doctrinal traditions. As a consequence, themonetary indemnities granted in the membercountries vary considerably in amount and cannotbe explained simply by a question of standard ofliving. Table 1 shows some examples of averageindemnity for injury in different countries.

EVALUATION SYSTEMS IN EUROPE

Leaving aside the problems in respect of howthe scale works, which we will address later, whathas given rise to the reform has been the obligationfrom the European Union, via the Fifth Directive,to reinforce the protection of those injured in roadtraffic accidents, as well as to speed up thesettlement of indemnities. For this reason, and inour opinion, we consider it very necessary to fullyunderstand the different systems for evaluatingbodily injury in the rest of the EU and, thus, totake them into account for the future reform of theSpanish scale.

In the financial evaluation of bodily injuryarising from traffic accidents, there are importantdifferences between the models applied in thedifferent European countries. On the one hand,countries such as Belgium, France, Italy and Spainapply models based on medical scales for theevaluation of the injury within their judicialsolutions. On the other hand, other countries likeIreland, Germany or England, in line with theirlegal tradition, have established judicial scales. InBermúdez, Ayuso and Santolino (2009) we set outthe characteristics of the evaluation systems of thesecountries and of others such as Sweden or Norway.

TABLE 1Differences in the monetary indemnity for non-pecuniary damages (€)

Austria

Belgium

England

France

Germany

Italy

Netherlands

Cuadriplegia

110.000

125.000-237.000

260.000-333.000

270.000

175.000-200.000

390.000

55.000-95.000

Masculine sexualmalfunction

47.000

12.000-50.000

127.000

76.000

15.000-75.000

115.000

35.000

Loss of vision inboth eyes

100.000

60.000-100.000

215.000

227.000

75-250.000

372.000

46.000

Loss of the sense oftaste

10.000

15.000

32.000

-

30.000

9.000

5.000-15.000

Source: Rogers, W.H. (ed.) (2001) Damages for Non-pecuniary Loss in a Comparative Perspective, Springer: Wien / New York.


should take into account that this European scale isonly a medical scale and, therefore, commonmethodology is required that will translate thepsychophysical damage suffered by the victim intomonetary terms. Undoubtedly, this second step inthe harmonisation process could be very complex,since the chosen methodology has to take intoaccount the economic differences in the differentstandards of living in the EU countries.

Nevertheless, with regard to a future reformof the Spanish scale, the European medical scaleproject should be taken in consideration whenanalysing the possible changes to the current TableV. On this point, the European scale establishes amore precise terminology as to what should beevaluated and certain norms or more exhaustiverules of usage to facilitate the work of the medicalexpert, which has been one of the complaints ofthe Spanish scale.

OUTLOOK ON THE REFORM OF THE SCALE

After more than a decade´s experience in theapplication of the current indemnity evaluationsystem, the deficiencies have become apparent, aswell as the need to update it so that it is in linewith present Spanish socio-economic situation. Webelieve that a legal reform of the scale ofindemnities should address judicial questions,medical criteria and socio-economic aspects.

First of all, the legal aspects of the reform ofthe scale are being debated extensively in specialistreading1 and in the judicial committees created bythe different institutions linked to the reform.Aspects from the most formal and structural to themost fundamental, such as the binding nature,

The EU has to facilitate the free movement ofpersons within the market and, considering thatthere is more and more cross-border litigation inrespect of road traffic accidents, it should alsopromote the harmonization of evaluation practicesapplied by the different member states. The firststep in achieving the aforementionedharmonization is the project entitled «Europeanscale Guide for the evaluation of the physical andpsychological injuries» already passed by theCommission at the end of 2003 (Rothley, 2003).Borobia (2006), one of the working parties set upto create the Guide, warns that the objective of thepublication of a European scale is to provide lossadjusters of the EU member countries with a guideon the grade of disability which arises from aspecific injury and, in this way, ensure a certaincoherence between the indemnity basis used in onecountry or another. It is currently in the process ofbeing approved in the European Parliament whereit will presumably be published as aRecommendation or as a Directive for road trafficaccidents that take place outside the State of theinjured person. As from May, 2005, this scale is nowobligatory for the evaluation of the seriousness ofinjury insured under health or personal accidentpolicies or workman’s compensation in Europeaninstitutions.

Despite the advance represented by thisproject for the harmonization of the indemnitiesfor bodily injuries between the member states, we

1 For example: Medina (2007), Vicente (2007),

Fernández (2008) y Xiol (2008).

➜THE EUROPEAN UNION URGED, VIA THE FIFTH DIRECTIVE, TO REINFORCE THE PROTECTION OF THOSEINJURED IN ROAD TRAFFIC ACCIDENTS, AS WELL AS TO SPEED UP THE SETTLEMENT OF THEINDEMNITIES


with the principle of integral repair. In most cases,it is impossible to know precisely the amount offuture financial loss and, therefore, it is impossibleto know whether the principle has been observed.The future reform of the scale will have to tackle,as far as it is possible, the apparent contradictionbetween the principle of integral repair and theprinciples of predictability, moderation andequality that it inspires.

Secondly, as a system of evaluation of personalinjury, those aspects related to the medicalevaluation represent one of the key points of thescale. Nevertheless, except by way of the commentsmade in the previous paragraph on the problemsdetected and the possible use of the Europeanrating scale project, we will not analyse further thisimportant aspect since we do not possess thenecessary knowledge to do so.

Finally, from a socio-economic point of view,the reform of the scale must improve theprotection of victims of road traffic accidents,following the guidelines of the Fifth Directive. Inthis sense, the reform of the scale should achieve asimilarity with the average levels of indemnity ofthe EU countries. Compared to these countries,the average level of indemnity derived from theapplication of the current Spanish scale is lower,

concurrence and compatibility or universitalityhave been discussed, amongst others.

However, the most outstanding points alludeto the basic principles of legal liability: theprinciple of vertebrate and integral repair. Asgeneral rule, we believe that the future reform ofthe scale should represent progress in the fulfilmentof these two principles, without forgetting theobjectives of predictability, moderation andequality on which the current system is based, aswell enforcing its binding character whichguarantees that the aforementioned targets are met.

In order to comply to a greater degree withthe principle of vertebrate repair, the future scalewould have to prevent specifying the same amountto repair simultaneously damages of a differingnature, as happens at present. In effect, the reformof the scale should look to breakdown theindemnities, differentiating between the amountsthat refer to pain and suffering and those damagesfor financial loss. At the same, damages forfinancial loss should be divided between thoseamounts paid for damages already incurred and forthose that correspond to future financial loss.

The latter concept, the future financial loss,which is difficult to quantify, makes the currentsystem questionable with regard to compliance


claims in motor accidents. In the followingparagraph, by way of a proposal, we show differentpossibilities to value the financial impact of thedifferent alternatives that could be put forth duringthe reform of the scale process.

EVALUATION OF THE SOCIO-ECONOMICIMPACT OF A POSSIBLE REFORM OF

THE EVALUATION OF PERMANENT INJURIES

In order to undertake evaluations of theeconomic impact of possible reform scenarios, it isnecessary to have a wide database on road trafficaccident claims involving victims that have sufferedbodily injuries. As an example, in our case, we hadinformation on a total of 22,709 motor claims thatoccurred in Spain between 2000 and 2007 and inwhich there was at least one victim that sufferedbodily injury (cases involving the death of thevictim were not included). The information refersessentially to injuries (temporary and permanent)suffered by individuals, including the consequences,the days off-work and the indemnities paid for eachtype of injury. However, it also includes relevantinformation on the characteristics of the claim, suchas the time taken to settle or where the accidenttook place.

If our objective is to value the economicimpact on the total cost of settlement of a reformproposal, first of all we have to ascertain thestructure of the aforementioned cost. Thebreakdown by type, for all of the claims, taking intoaccount the information of the sample study, isshown in figure 1.

In this first approach to the problem of valuing

and clearly lower, in those claims with seriousinjuries. After almost fifteen years of functioning,the current scale needs a reform that is able tocompensate more fairly those who suffer seriousinjuries, in line with European levels.

Nevertheless, although there is a certainconsensus to improve the awards for seriousinjuries along the lines of our neighbouringcountries, we believe that there has been littledebate over the socio-economic impact that animprovement such as this could represent on thetotal cost of indemnities to be paid by insurancecompanies. Moreover, we must bear in mind thatalthough initially it would be the insurancecompanies that would suffer the problems relatedto the management of this improvement, it wouldbe society as a whole (or at least the vehiclesdrivers) that would have to bear the increase of theoverall cost, possibly via an increase in thecorresponding insurance premiums.

This socio-economic component should beborne very much in mind throughout the wholereform process and, in our opinion, very much as akey matter and not just a collateral one. For thisreason, each proposal for the reform of theevaluation system should have its own study on theeconomic impact that it could represent. In thisway, the difficult process of choosing between thedifferent proposals made would have more andbetter elements to be considered.

In Bermúdez, Ayuso and Santolino (2009) weconsidered different reform scenarios and evaluatedtheir impact on the total cost of indemnitiesthrough the use of a database of personal injury

➜THE REFORM OF THE SCALE IN SPAIN SHOULD ACHIEVE A SIMILARITY WITH THE AVERAGE LEVELS OFINDEMNITY OF THE EUROPEAN UNION COUNTRIES


the economic impact of different proposals for thereform of the scale, interest has been focused on thepart of the system related to permanent injuries. Inthis sense, the reform scenarios considered hererefer to the modification of aspects contained inTable III and IV of the current scale. The proposedscale reform scenarios, where the impact on thetotal settlement cost has tried to be estimated, havebeen formulated bearing in mind the opinions onthe reform from the various articles, documents andforums analysed, including our own evaluation.

guidelines towards greater protection of the victim setout under the Fifth Directive. Also, in view of the cle-ar shortcoming in the level of indemnity for se-riously injured persons, we considered that the incre-ases in the monetary value point should be higherwhen the overall score is higher. Following theseconsiderations, as can be seen in Table 2, we establish

First of all, we propose a series of scenarios inrelation to a future modification of the monetarypoint value in the current Table III. To define thesescenarios, we started out by considering that themonetary point should be increased following the

➜THE REFORM OF THE SCALE SHOULD FOLLOW THE OBJECTIVES OF FORESEEABILITY, MODERATION ANDEQUALITY THAT THE CURRENT SCALE WAS BASED ON, AS WELL AS MAINTAINING ITS BINDING NATUREIN ORDER TO GUARANTEE THE ACHIEVEMENT OF THESE TARGETS

27%

4%

37%

14%

2%

14%2% Indem permanent injury

Indem aesthetic damages

Indem temporary injuries

Correction factor perm injur

Correction factor temp injur

Hospital & legal expenses

Legal and delay interests

FigurE 1. Breakdown of total cost for permanent and termporary injuries.


five score ranges and, for every scenario, we took dif-ferent percentage increases in the point value foreach of the defined range. In all cases, the increase ishigher in proportion to the increase in the injury af-ter-effects.

Having defined the scenarios and with the helpof the database which indicates the points awarded toeach victim for physiological and aesthetic damagesfrom the application of the medical scale, calculationscan be carried out to ascertain how the overall in-demnities will be increased for injury after-effects (orbasic indemnity). For example, for scenario 2, the in-creases proposed in Table III, of up to 25 per cent forserious injuries, would produce an increase in the in-demnities for after-effects of 6,19 per cent, which atthe same time would represent an increase of 1,96 percent on the total settlement cost.

Secondly, we wished to look at different reformscenarios in relation to the concepts of the currentTable IV (corrective factors that includecompensations for financial damages, permanentdisability, the need for assistance from anotherperson, adequacy of housing and of own vehicle).According to the available sample, more than onethird of the victims have not received compensation

with corrective factors, and 60 per cent of themhave received compensation for financial damages.Hence, only 4 per cent of victims receivedcompensation different to that of that of financialdamages. However, this small percentage representsmore than 80 per cent of the total amount settled ascorrective factors. The amounts for permanentdisability and assistance from another person are

TABLE 2

Evaluation of the economic impact in the case of a modification to themonetary point value for physiological and aesthetic damages

Aesthetic and after-effect damages

Increase (%) for range 1-10





%increase on basic indemnity

% increase on total cost

Scenario 1

0

0

5

10

20

4,44

1,40

Scenario 2

0

5

10

15

25

6,19

1,96

Scenario 3

5

10

20

30

40

13,60

4,30


much higher than those related to compensationsfor financial damages. For these reasons, whenlooking at reform scenarios and evaluating itsimpact on the overall settlement cost, we shouldconcentrate on the three most representative groups:financial damages, permanent disabilities and needfor assistance by another person.

As we indicated in the previous paragraph, theevaluation of future financial damages represents oneof the most delicate points within the future reformof the scale. At present, the compensation of thefuture financial damages is limited by the maximumamounts shown in Table IV for each of thecategories of permanent disability (partial, total andabsolute). This limitation is questioned by mostspecialist lawyers who think that for an integralrepair of the future financial damage, in the contextof a closed and binding system such as ours, it isnecessary to apply a series of «presumed» rules.These rules have not yet been treated by thespecialists that are working on the future reform ofthe scale, although those used in English legislation(GAD, 2007) could be used as a reference.

With a view to undertaking the first evaluationon the economic impact of the possible reforms thatcould be proposed on this point and withoutdiscussing the necessary assumptions for obtaining«presumed» rules that provide maximum guaranteeson the principal of integral repair, we haveconsidered that these items, together with thoseothers already referred to, must give rise toindemnities resulting from the application of theevaluation scale for personal injuries from roadtraffic accidents that are comparable with theindemnities paid in the EU. Let's consider, forexample, an extremely serious case. It involves a 20-year-old who is tetraplegic due to an accident andwhose income was 30.000 euros a year.

In Table 3, in the first column, the possiblemaximum indemnity appears according to the 20072

for the case described3 and, in the following

columns, three scenarios presented to obtainmaximum indemnities of two, two and a half andthree million Euros, respectively. For these scenarios,the same assumptions have been made for the threeitems except that of permanent disability. In the firstcase, an increase of 25 per cent on the point value asshown in scenario 2 of Table 2. Secondly, foreconomic damages, there is an increase on the basicindemnity of 15 per cent and, in the third case, forthe indemnity due to the need for assistance ofanother person, a maximum indemnity has beenconsidered equal to the actuarial evaluation of themaximum revenue to be earned by a 20-year-oldage victim with severe dependency (level II) inaccordance according to the Dependency Law4.Finally, the remaining items have been maintainedwith the maximum indemnity as per the 2007 scale.

2 The 2007 scale has been used for the purposes

of comparing the sample data available.3 It can be seen that the maximum possible

indemnity as for the 2007 scale is around

1.200.000 euros. In other EU countries the

indemnity for a similar victim goes from 7

million Euros in the United Kingdom to almost

4 million Euros in Germany.


If convergence towards European indemnitylevels is sought, then the evaluation of the futureeconomic damage must increase considerably as canbe appreciated from Table 3, observing the values inthe total and permanent incapability disability itemsto achieve the maximum total indemnities thatbring us closer to European levels.

Specifically, the maximum quantity of this itemshould be increased by 243,03 per cent to reach thetotal figure of two million Euros in the firstscenario, and 545,38 or 847,73 per cent to reach thefigures in scenarios 2 and 3, respectively. In view ofthese amounts, we need to consider the question ofthe impact that such an increase would have on theoverall settlement cost, keeping all other itemsunchanged. As is obvious, the increase in themaximum amount is higher than what would haveto be applied, on average, to the three types ofdisabilities. To calculate the increase to be applied tothe total costs for disabilities, we have consideredthat distribution of the increases is normal, that theincreases calculated for an extreme case are thedistribution maximum and, therefore, on average,the increases should be 50 per cent lower than themaximum. On reviewing Table 3, one can conclude

TABLE 3

Maximimum Total Indemnity Scenarios

Basic Indemnity (100 points)

Aesthetic Damages (50 points)

Financial Loss

Complementary Pain and Suffering damages

Total and permanent Disability

Need for 3rd party Assistance

Adaption of Home

Adaption of own vehicle

Total Maximum Indemnity

% on corrective factors total

% on total cost

2007 Scale

300.496,00

150.248,00

67.611,60

82.685,58

165.371,17

330.742,34

82.685,58

24.805,67

1.204.645,94

Scenario 1

375.620,00

187.810,00

84.514,50

82.685,58

567.276,67

594.602,00

82.685,58

24.805,67

2.000.000,00

66,32

9,09

Scenario 2

375.620,00

187.810,00

84.514,50

82.685,58

1.067.276,67

594.602,00

82.685,58

24.805,67

2.500.000,00

186,90

25,61

Scenario 3

375.620,00

187.810,00

84.514,50

82.685,58

1.567.276,67

594.602,00

82.685,58

24.805,67

3.000.000,00

269,59

36,95

4 From our point of view, the future reform

of the scale should compensate this concept

as per Law 39/2006, of 14th. December, for the

Promotion of Personal Autonomy and

Assistance for persons in a situation of

Dependency (commonly known as the

Dependency Law).

➜THE CREATION OF THE NEW SCALE SHOULD INVOLVE AN EQUAL PARTICIPATION OF EXPERTS FROM THEDIFFERENT DISCIPLINES IN ORDER TO AVOID SOME OF THE CONTRADICTIONS OF THE CURRENTSCALE


from the examples taken for scenario 2, thatcompensation of the future financial loss atEuropean levels would represent an increase ofapproximately 25 per cent in the total cost ofsettlement of personal damages.

Finally, we would point out that the cases ofevaluation presented here are simple examples as tohow we might proceed to obtain an economicreport on the impact of every proposal within theentire settlement cost. If we were to have moredetailed information on the different scale reformproposals, and with the help of a database similar tothe one that was made available, we believe that theresults that would be obtained, following proceduressimilar to those used, would be of great help intaking decisions and, to be exact, to evaluate thescope of the scale reform.

CONCLUSIONS

Finally, and in summary, we would emphasisethat it is generally agreed in economic circlesinvolved that there is a need to reform the existingscale and that, ultimately, the future reform mustaddress an improvement in the protection of roadtraffic accident victims. Also, we feel that thereform should follow the objectives offoreseeability, moderation and equality that thecurrent scale was based on, as well as maintainingits binding nature in order to guarantee theachievement of these targets.

From our point of view, the objectives thatinspired the current scale still prevail. A highdegree of predictability is necessary to facilitateagreements and to avoid, as far as possible, legal

disputes. In this way, the heavy litigious nature ofour judicial system would be reduced and theindemnification of the road traffic victims wouldbe accelerated. At the same time, theaforementioned predictability would facilitate theestimation of the technical reserves by insurers and,thus, increase their solvency.

The reform of the scale must continue tocomply with the constitutional principle ofequality, although sometimes it might be difficultto make this compatible with the objective ofpredictability. The factors and circumstancesrelating to a road traffic accident and its victims areso many and diverse that makes it impossible toreflect them all in a fixed table such as the scale. Abasic indemnity, which takes into account some ofthe more objective factors, guarantees the principleof equality. However, if we include correctioncriteria on the basic indemnity, which are necessaryto take into account the remaining factors, we will

➜THE REFORM OF THE SCALE MUST CONTINUE TO COMPLY WITH THE CONSTITUTIONAL PRINCIPLE OFEQUALITY, ALTOUGH SOMETIMES IT MIGHT BE DIFFICULT TO MAKE THIS COMPATIBLE WITH THEOBJECTIVE OF PREDICTABILITY


have reached a consensus or balance between thepredictability and equality objectives.

With regard to the objective of moderation,sometimes criticized by certain sectors whoconsider it to be a demand of the insurancecompanies´ lobby, we should remember that, priorto the approval of the current scale, the situationof judicial inflation in the indemnities seriouslyprejudiced the financial situation and solvency ofinsurers and hence their principal concern was tomoderate the aforementioned judicial inflation.Presently, and from our point of view, we shouldconsider this objective as a guarantee of the non-profit principle of insurance, i.e. a road trafficaccident victim cannot profit from the occurrenceof the accident and should only be compensatedfairly for the damages caused.

In order to achieve the set goals, we believethat the reform of the scale should take intoaccount a series of circumstances. First of all, areform of the system for evaluating personaldamages should respond to multidisciplinarycriteria, i.e. to legal, socio-economic and medicalcriteria. The creation of the new scale should

involve an equal participation of experts from thedifferent disciplines in order to avoid some of thecontradictions of the current scale. In this sense, aspointed out in previous paragraphs, it is necessaryto underline the need to have socio-economicreports for each proposal of reform of the scale, sothat the impact of the proposal can be evaluated inrelation to the total cost of indemnities for roadtraffic accidents.

Secondly, the implied financial partiesinvolved must respond to the challenges posed bythe reform of the scale. On the one hand, viadiscussion and reflexion the regulator-legislatorshould encourage agreements that guarantee afairer and more efficient evaluation system. On theother hand, the insurance companies, and as theyhave been doing, should participate in the debateof the reform and, thereafter, take onboard thechanges that it produces. Lastly, society in generalhas to understand that greater protection for thevictims will probably mean higher claimssettlement costs for insurers and, consequently,higher premiums for all policyholders. On thispoint we ask ourselves whether the insurancesector, in its widest sense and consisting ofpolicyholders, insurers and the regulator, is ready tobear an increase in premiums for the motor class ofup to 25 per cent as a consequence of the increasein the cost of indemnities for personal damages.Perhaps, the modification of the scale in phasescould be considered so that the increase in phasescould be gradual.

However, we believe that there are certainactions that could be taken to mitigate the effectthat an improvement in the protection of victimscan have on the premiums. The introduction of theconcepts of offer and motivated reply may give riseto a reduction in the settlement costs in respect ofinterests for delay or legal costs. Moreover, theexpenses related to the reimbursement of damagesand/or compensation (hospital expenses, attorney


and solicitors fees, etc.) might reduce as and whenagreements are reached with the relevantauthorities.

The future reform of the scale could favour aredistribution of the current compensations. Lessergenerosity in the indemnities for injured persons ofsmaller severity (very frequent) might «finance»substantial increases in the indemnities for moresevere injuries (not so frequent and currently worsecompensated under the scale), without representinga substantial increase of the premiums.

The latter point might end up being crucialbut undoubtedly will not be exempt fromcontroversy. It is our opinion that we should takeadvantage of the reform of the scale to moderatethe amount of indemnities for minor injuries. Ahigh percentage of these are in relation to cervicalwhip lash which, in our understanding, areindemnified in excess of what we consider to be afair amount or are simply fraudulent claims in viewof the difficulty in proving these injuries.

Finally, we do not want to close this workwithout sending out a message of optimism in viewof the effort being made by Spanish society toreduce the accident level on our roads and which isthe best solution to the endemic problem of bodilyinjuries in road traffic accidents. ❘

REFERENCES

BERMÚDEZ, L.; AYUSO, M.; SANTOLINO,M.Outlook and economic analysis of the futurereform of the system for evaluating bodilyinjury. Madrid: FUNDACIÓN MAPFRE,Cuadernos de la Fundación nº 145, 2009.BOROBIA FERNÁNDEZ, C. Evaluation of bodilyinjury: Legislation, methodology and expertmedical proof. Barcelona: Masson, 2006.FERNÁNDEZ MARTÍN, M.J. The modification ofthe regulation on insurance of legal liability formotor vehicles. Revista Española de Seguros,2008, nº 136, 427-478. GAD. Actuarial tables with explanatory notes foruse in Personal Injury and Fatal Accident Cases.London: 6th edition, Government Actuary’sDepartment, The Stationery Office, 2007.Medina Curl, M. Solid bases for a conservativereform of the legal evaluative system.. RevistaEspañola de Seguros, 2007, nº 131, , 271-296.ROTHLEY, W. Draft report withrecommendations to the Commission on toEuropean disability rating scale. LegalCommittee on Affairs and the Internal Market,2003/2130 (INI), 2003.VICENTE DOMINGO, E. More than a decadewith the Scale: Analysis of the Supreme Courtjurisprudence in the context of STC181/2000, of June 29: a reform proposal.Revista Español de Seguros, 2007, nº 129-130,171-204. XIOL QUINGLES, J.A. The motivated offer ofindemnity for insurers under Law 21/2007 ofJuly 11, for reform on legal liability andinsurance for motor vehicles. Revista Españolade Seguros, 2009, nº 137, 59-82. XIOL RIVERS, J.A. Need for a change in theEvaluation System of bodily injuries. In: XVICongress of Legal liability, 6-7 March,Barcelona, 2008.


The practical application ➜of credibility

In an increasingly competitive market, fixing

insurance premiums is becoming an essential

task for insurance companies. The theory of

credibility, using its models, enables us to use

our own claims experience as basic information

for calculating the premium of a specific

insurance. By means of a practical example of

Health Insurance, this study attempts to

demonstrate how the application of credibility

models is the best method to adjust the premium

to the real claims experience of the product.

INMACULADA PEÑA SÁNCHEZMAPFRE FAMILIAR

models in the rating of HealthInsurance


starting points they all converge into the sameresult. They all suggest that the premium to bepaid by the insured should combine theindividual experience (of the insured group)with that of the product in general so that apremium is obtained that can guarantee theprinciples of sufficiency and fairness.

CREDIBILITY MODELS

This study focuses on the two mostimportant classic models; the Bühlmann andthe Bühlmann-Straub free distributionmodels. They are named free distributionmodels because they do not need to establishhypothesis either on the distribution of theindividual risks nor on the distribution apriori of the risk parameters.

Whatever the credibility model used, it isnecessary to previously define the initial data,specifically:

■ k, inured groups with homogeneous,but not identical, characteristics. Namely,there should be a certain independencebetween the groups. In this study, thegroups of insureds have been formed inage bands.■ n, number of periods of observationwhich, in this case, is a year.■ Xij, is the principal random variablethat includes the overall individualclaims experience of the j-th in year i.■ wij

1 is the observed weighting factor.A weight is assigned to each of the Xij.

In an increasingly competitive market,fixing insurance premiums is becomingan essential task for insurance companies.The greater the knowledge of the risk to becovered, the more exact the insurancepremium will be.

The theory of credibility, using itsmodels, enables us to use our own claimsexperience as basic information forcalculating the premium of a specificinsurance.

By means of a practical example ofHealth Insurance, this study attempts todemonstrate how the application ofcredibility models is the best method to adjustthe premium to the real claims experience ofthe product.

CREDIBILITY THEORY

The credibility theory consists of a seriesof statistical skills aimed to calculate theinsurance premium based on the individualclaims experience of the product.

The application of this theory is basedon the rating, a posteriori, and which attemptsto the heterogeneity within each risk factor.For example, in the case of a product withseveral insured groups, each group will have aparticular claims experience over time. If onegroup has a high claims experience, it islogical to penalize this group, and not all ofthem.

The evolution of this theory over timehas given rise to different credibility models,with the peculiarity that taking different

THE CREDIBILITY

THEORY

CONSISTS OF A

SERIES OF

STATISTICAL

SKILLS AIMED TO

CALCULATE THE

INSURANCE

PREMIUM BASED

ON THE

INDIVIDUAL

CLAIMS

EXPERIENCE OF

THE PRODUCT

«The essence of wisdom is concern for the future» Russel L. Ackoff

1 Applicable only to the Bühlmann-Straub model.


■ m, group estimator. It is the riskpremium or the expected value for all ofthe individual risk premiums. ■ a, indicator of the heterogeneity ofthe portfolio being analyzed.. ■ S2, overall measurement of dispersion

After applying the credibility models, aseries of parameters are obtained which willhelp me analyse the results obtained for eachmodel.

The principal structural parameters areas follows:

ILLU

STR

ATIO

N S

TOC

K


the credibility factor with the difference that,in this model, each of the groups has its ownzj credibility factor.

The credibility premium estimator inthe Bühlmann-Straub method is shown in thefollowing formula:

µ(θj)= (1-zj)· m + zj · xjw

PRACTICAL APPLICATION

The objective of this application is noneother than to see, in practical terms, how todetermine the best linear premium byapplying the credibility models to a portfolioof a Health product.

The results obtained will be compared tothe company´s current method for calculatingpremiums in order to analyze whether or notthey coincide with the actual business.

We will now analyze in detail eachstage in the application of the credibilitymodels.

1. Selecting the portfolio for studyThe study is applied to an Individual

Health Care Insurance Product. It is definedas an insurance contract by which thecompany, after payment of the premium bythe insured and through its facilities,undertakes the commitment to provide thenecessary medical, surgical and hospitable carein the event of illness or injury.

With a view to selecting this product,prior analysis was carried out by age and sexon the evolution of the exposure2 in time.

for the claims experience over time. ■ z, credibility factor. It is the degree ofconfidence given to the individualclaims experience of the portfolio.

Bühlmann modelThe Bühlmann model was really the first

credibility model in modern theory, and onwhich Theory of the Credibility is based

Bühlmann estimates the free distributioncredibility formula based on the minimumsquares criteria. The risk premium is afunction of the credibility factor z which is alinear combination of both the specific andgeneral information. In this model, thecredibility factor is the sole factor for thewhole portfolio.

The credibility premium estimator forBühlmann credibility is shown in thefollowing formula:

µ(θj)= (1-z)· m + z · xj

Bühlmann-Straub modelThis is an amplification of the Bühlmann

model using an additional variable, the knownweighting factor wij. This way, each Xijobservation of the model will have a wijassociated weight. With this variable, thehomogeneity of the observations is eliminatedover time; a characteristic that prevails in theBühlmann model.

The Bühlmann-Straub approach alsoestimates the risk premium as a function of

THE STUDY

FOCUSES ON THE

TWO MOST

IMPORTANT

CLASSIC

MODELS: THE

BÜHLMANN AND

THE BÜHLMANN-

STRAUB FREE

DISTRIBUTION

MODELS

2 All units that generate risk during the period considered are deemed to be exposed.

23

Source: prepared by author.


Graph1. Evolution of the individual Health Care exposure by gender.

0 10 20 30 40 50 60

2005 2006 2007 2008

AGE

2009

EXP

OS

ED

0 10 20 30 40 50 60

2005 2006 2007 2008

AGE

2009

EXP

OS

ED

Individual Health Care exposure. Women

Individual Health Care exposure. Men

24

● As it has the highest weighting on the premium and,● Because it presents a greater frequency compared to other covers.

The considered observation periodstake into account the years of claimsexperience in the company´s possession.This study considers the periods from 2005to 2008, making the analysis for the 2009year.

Comparing both graphs the followingconclusions can be made:

–The exposure presents a similarbehaviour both in men and in women,showing the highest values between theages of 30 and 40.– The number of those exposedincreases progressively with age, and itis from the age of 18 when the numberof women surpasses that of men.

2. Claims experience informationHealth Care insurance includes several

covers, such as primary care, specialistattention and obstetric assistance, to namethe more important.

In order to use a solid claimsexperience and sufficient volume on whichto apply the models, the Specialist Assistance3

cover has been taken for the following tworeasons:

THE THEORY OF

CREDIBILITY,

USING ITS

MODELS,

ENABLES US TO

USE OUR OWN

CLAIMS

EXPERIENCE AS

BASIC

INFORMATION

FOR

CALCULATING

THE PREMIUM

OF A SPECIFIC

INSURANCE

600

500

400

300

200

100

0

0 10 20 30 40 50 60

2005 2006 2007 2008

AGE

UN

IT C

OS

T


Graph 2. Evolution of the unit cost in Specialist Assistance by gender.

3 Specialized Assistance includes specialist medical consultations as well as the tests carried out but does notinclude the obstetric speciality.

Claim unit cost Specialist Assistance. Men


the unit cost during all the consideredobservation periods, each unit cost prior tothe point of analysis, the 2009 year, will beupdated in accordance with the indicationfrom the INE (National Statistics Institute)for medical and similar services, as reflectedin the following graphs.

By analyzing in detail the behaviour ofthe unit cost, it can be seen graphically that:

–The evolution of the unit costincreases progressively with the age,both for men and for women, althoughthe growth is even more pronouncedfor women. With the more advancedages, there is a greater dispersion in thedata.–The amount of the unit cost is greaterin women than in men. From the age of20, women have twice the cost for men.However, men maintain a fairly stable

The variables in the credibility modelsare represented both by the unit cost of theclaim and the relative frequency.

Unit cost per exposureThe unit cost refers to the expense that

the insured incurs for any medicalconsultation as well as for the medical teststhat are necessary during the time that theinsured remains in policy. This informationwill be what is shown in the claimsexperience observations as the principalvariable in the credibility models.

In order to compare the evolution of

600

500

400

300

200

100

0

0 10 20 30 40 50 60

2005 2006 2007 2008

AGE

UN

IT C

OS

T


IN THIS STUDY,

THE VARIABLES

IN THE

CREDIBILITY

MODELS ARE

REPRESENTED

BY THE UNIT

COST OF THE

CLAIM AND THE

RELATIVE

FREQUENCY

Claim unit cost Specialist Assistance. Women


20

15

10

5

0

0 10 20 30 40 50 60

2005 2006 2007 2008

AGE

FR

EQ

UE

NC

Y

20

15

10

5

0

0 10 20 30 40 50 60

2005 2006 2007 2008

AGE

FR

EQ

UE

NC

Y

Graph 3.Evolution of the relative frequency in Specialist Assistance by gender.



Relative frequency in Specialist. Men

Relative frequency in Specialist. Women


3. Definition of the starting pointhypothesis

Once all the variables of the modelhave been defined, the starting pointhypotheses can be established for thisapplication:

● Two different models depending on the gender, one for men and one for women, since each answers to a different behaviour of the unit cost and frequency.● Principal variable of risk model, Xij, the unit cost.● Additional variable, the weighting factor wij

4 the relative frequency. ● Insured Groups, k. In view of the need to create homogeneous but not identical groups, both the exposure and the unit cost by age and sex have been analysed, considering the deviation from the average. In this way, groups have been created by age as a function of cost and number of exposures in order to maintain a certain independence and homogeity.● Observation periods, n. 4 years, from 2005 to 2008.

4. Application of the credibility modelsOnce the initial hypotheses have been

established and all the individual claimsexperience information has been prepared inaccordance with the variables defined in themodels, unit cost and frequency, theBühlmann and Bühlmann-Straub credibilitymodels can be applied.

The result of the application of themodels will produce the lineal credibilitypremium which uses both the individual

unit cost up to the age of 45 when thecost begins to increase significantly.

Relative frequencyRelative frequency is the name given to

the number of claims per exposure, a claimbeing the performance of a medical act. Withthe frequency data a weight is assigned toeach of the unit cost observations.

As can be observed in the graphs, theevolution of the frequency during the yearsconsidered is as follows:

–The behaviour of the frequencyfollows a trend similar to that of theunit cost.–The evolution of the frequencyincreases progressively with age and is

THE CREDIBILITY

THEORY

THROUGH

CREDIBILITY

MODELS OFFERS

THE CLEAR

ADVANTAGE OF

PROVIDING A

PREMIUM

ADJUSTED TO

THE REALITY OF

THE PORTFOLIO

more pronounced in the case of thewomen. As in the unit cost, with moreadvanced ages, the dispersion in thedata increases.–There is a clear difference in thefrequency by gender. In men, thefrequency remains almost constant upto the age of 40, when it begins toincrease. However, in the case ofwomen the increase can already be seenfrom the age of 15. 4 Applicable only to the Bühlmann-Straub model.


b) Credibility factor, z:It will use values between 0 and 1.Thecloser the credibility factor is to one, thegreater the weight that the individualinformation will have on the credibilitypremium.In this case, z is very close to one in bothmodels although slightly higher in theBühlmann model.This will mean that theweight that the individual information hason the credibility premium will be greaterthan the global information.

models similar values are obtained.● m, portfolio heterogeneity indicator.The greater the heterogeneity of thegroups within the portfolio, the greaterthe weight of the individual experienceprovided. It can be seen that in bothmodels the heterogeneity is fairly high.● S2, global dispersion measurement ofthe individual claims experience.Thelesser the dispersion, the lesser therandom quality of the information andthe individual experience will have agreater weight.A minimal dispersion isobtained in both models that impliesgreater credibility of the individualinformation provided.

experience as per the credibility factor (z) aswell as the overall experience, (1-z), reflectedin the following formula:

Credibility risk premium = z· Xj + (1– z)· m

5. Analysis of resultsAnalysis criteriaThe measurements that are used to

select the credibility model which bestreflect the reality of the business are thefollowing:

a) Structural parameters:● m, group estimator. It is premium resultusing only the overall experience. In both

THE CREDIBILITY

FACTOR “Z” IS

THE DEGREE OF

CONFIDENCE

GIVEN TO THE

INDIVIDUAL

CLAIMS

EXPERIENCE OF

THE PORTFOLIO

Parameters

m

a

s2

Gender

MF

MHF

MF

Bühlmann

147,98209,54

4.433,129.830,27

41,6459,87

Bühlmann-Straub

148,09209,81

5.930,329.004,37

365,31796,47

Table 1. Results of the structural parameters of the credibility models.

Source: prepared by author


similar to the individual estimator Xj thanthe group estimator m.

Comparing both credibility models, itcan be seen that premium of the Bühlmannmodel is lower.

c) Credibility premium:As a direct result of obtaining

credibility near to one in both models, thecredibility premium obtained will be more

Group

1234567

137,1991,6393,54107,30136,91192,13277,13

Individual estimatorXj

123

147,98567

Group estimator m

123

0,9977567

Credibility factor z

137,2291,7793,66107,40136,94192,03276,83

Credibility riskfactor

Table 2. Results of the Bühlmann model.

Group

1234567

119,6785,83

144,18213,39243,79305,20354,74


123

209,54567

Group estimatorm

123

0,9985567

Credibility factorz

119,8186,02144,28213,38243,74305,06355,37


Table 4. Results of the Bühlmann model.

Group

1234567

137,0991,6393,55

107,32136,93192,28277,52


123

148,09567

Group estimatorm

0,99680,99650,99650,99680,99740,99810,9986

Credibility factorz

137,1391,8293,75107,45136,96192,19277,34


Table 3. Results of the Bühlmann-Straub model.

Men

Women



where it reaches the highest values.Although an additional variable, relative

frequency, wij, is included in the Bühlmann-Straub model, these frequency values arehomogeneous in time and, therefore, neitherenrich the model nor improve the results ofthe Bühlmann model.

Hence, the model that most favours theuse of the company´s own experience is theBühlmann credibility model.

7. Adjustment of the credibility modelswith the current method

A simple way to verify whether thepremium currently applied by the companyresponds to the real claims experience of thebusiness is to compare it with the credibilitypremium obtained via the credibility model.

It should be clarified that, to adjust thecredibility premium to the business reality, itis necessary to calibrate the credibility modelaccording to the peculiarities of eachbusiness.

Comparison of premiums before the adjustmentA first comparison is made between the

company´s risk premium and the Bühlmannmodel´s credibility estimator which respondsto the unit cost.

6. Choosing the optimum modelBoth credibility models fulfil the

requisites of a sound credibility model. Thestructural parameters and the credibilityfactor demonstrate that, through thecredibility models, the individualinformation used is sufficiently reliable onwhich to base the risk premium calculation.

From the point of view of selecting theoptimum credibility model, and despiteobtaining very good results with bothmodels, there are some small differencesbetween the two:

–The dispersion (S2), although minimal,is even lower in the Bühlmann model.–The credibility (z) is very close to one,but it is in the model of Bühlmann

THROUGH

CREDIBILITY

MODELS, THE

INDIVIDUAL

INFORMATION

USED IS

SUFFICIENTLY

RELIABLE TO

BASE THE RISK

PREMIUM

CALCULATION

Group

1234567

119,6785,86

144,19213,40243,85305,24355,58


123

209,81567

Group estimatorm

0,99480,99470,99680,99770,99800,99840,9986

Credibility factorz

120,1486,53144,39213,39243,78305,09355,37


Table 5. Results of the Bühlmann-Straub.

: prepared by Source author


600

500

400

300

200

100

0

0 14 23 39 49 59 65

Bühlmann Current method

AGE

RIS

K P

RE

MIU

M

600

500

400

300

200

100

0

0 14 23 39 49 59 65

Bühlmann Current Method

AGE

RIS

K P

RE

MIU

M

Graph 4. Comparison of premiums before the adjustment.



Credibility Model vs Current method. Men.

Credibility Model vs Current method. Women.


Comparison of premiums after the adjustmentAt present, with the premium applied

by the Company, a claims ratio of 58% isbeing obtained. However, the credibilitymodel calculates a premium that responds tothe actual unit cost of the claim, i.e.,premium and claim coincide, that implies aclaims ratio of 100%. For this reason, theBühlmann model premium is lower than thepremium calculated by the company´scurrent method.

Since the claims ratio is the quotient ofthe claims amount divided by premium, thisimplies that the premium is inverselyproportional to the claims experience ratio.

Nevertheless, in order to make ahomogeneous comparison between bothmethods, the credibility premium mustrespond to the same claims ratio presentedby the company. If the claims ratio for the

From these graphs one can see that: –The company´s risk premium has anevolution by age and sex similar to thatobtained in the credibility model. –It can be clearly appreciated that thecurrent method used by the Companyhas a higher premium than thatobtained in the credibility model.

600

500

400

300

200

100

0

0 14 23 39 49 59 65


AGE

RIS

K P

RE

MIU

MGraph 5. Comparison of premiums after the adjustment.


Credibility model adjustment vs Current model. Men.


company in particular bases its premiumcalculations on its own claims experience.This signifies a greater knowledge of therisks to be covered and, at the same time,enables it to be competitive in the insurancebusiness. ❘

BIBLIOGRAPHY

GÓMEZ DÉNIZ, E; SARABIA ALEGRÍA,J.M. Teoría de la credibilidad: Develop-ment and Applications for Insurance Pre-miums and Operational Risks. Madrid:FUNDACIÓN MAPFRE, 2008.PEÑA SÁNCHEZ, I. Rating methods withcredibility in group insurances. ICEAPublication, February 2006, documentnº 185, 71-94.IBARRA ALFARAZ, J.A.; MONROBEL AL-CÁNTARA, J. R.I Rating and credibilitytheory.

Bühlmann credibility model reduces to 58 %that corresponds to the ratio obtained by thecompany, then the premium will increaseproportionally.

In this way, one can see graphically howthe Bühlmann model premium is very closeto the company´s premium, both for menand for women.

CONCLUSIONSThe credibility theory through

credibility models offers the clear advantageof providing a premium adjusted to thereality of the portfolio, charging premium inaccordance with the amount of risk,rewarding the good behaviour for those riskswith low claims experience and penalizingthose that present a poorer experience.

By way of the result obtained in thisapplication, the credibility models confirmthat the current method used by this

600

500

400

300

200

100

0

0 14 23 39 49 59 65


AGE

RIS

K P

RE

MIU

M


Credibility model adjustment vs Current model. Women.


FRANÇOIS SETTEMBRINORisk Manager FERMA

Every story, we are told, has a beginning and an

end. Open any paper, however, and you’ll see this

is not so. Take the Fortis case; only a very shrewd

observer could say when the shenanigans began.

Even shrewder he who could say when everything

will end. Much the same goes for all the other

shady customers who hog the headlines lately.

How did all that begin? How and when will it all

end?

The three a story withoutbeginning

or end

Rs;

The three Rs simply stand for the three greatchallenges we face today, namely risk,respect and responsibility. We have quitesimply forgotten them Let’s bring them back centrestage.


R R R

g


ahead. Has risk suddenly changed its spots? No, it isreturning to what it has always been, the only wayahead. Since mankind has been on the earth he hasalways had to run a whole gauntlet of risks, involvinghis peers, his environment or natural or artificialdisasters… But he has never been able to conceivethem as anything else than danger or damage ratherthan the other way about. The only rational conclusionwe can draw from all this is that risk is nothing morethan the variance of probabilities, for good or ill. Badtimes may be followed by better; businessmen, like all ofus human beings, always hope for the best. But a certainlucidity should make us aware that a hitch might alwaysbe in the offing, and we should never lose sight of thetwo faces of risk. Optimists will see the glass half fulland pessimists half empty. This different viewpoint doesnot change the actual event one whit: good and badwill always be bound up with each other. RiskManagement can only aim to fine tune this perceptionand increase our chances of coming out with a positiveoutcome. Risk Management is a factor of progress andin no way the harbinger of catastrophes, losses anddamage. That being so, how should we react? Whatwould be the best way of rekindling a risk culture tokick-start business anew? A good start would be toconvince managers that the situation is not reallycomplicated at all. In fact the truth resides in a simpleperception within reach of everyone; to strive for abetter world we have to run risks. This takes absolutelynothing away from all the procedures that have beendreamed up and developed hitherto: mathematical,actuarial, technical or legal. They are there to help thedecision taking but can never override the always

It is curious how many errors this little wordengenders. Many regard it as a synonym of dangeror damage. Few Risk Management handbooksshed any more light on the matter. Understood as such,it would always be worthwhile making an effort to skirtit. We are in fact besieged with advice of this tenor. «Asfar as possible we should avoid risks; if there is any wayof not taking risk this is the path we should take». If itcannot be completely eschewed, the best option is thento reduce it, trim it down by all possible measures ofprevention and protection. There is another way ofgetting out of the fix, having someone else run the risk;many business contracts do just that. They are the dailybread and butter of jurists and lawyers, and this allcomes out very expensive. If this solution is still reallyunsatisfactory, where else do we turn? Pass on the riskto someone else, i.e., an insurer. The latter will onlyagree to cover the damage or risk on certainconditions; the damage has to be quantifiable, arisingfrom acceptable and, above all, random risks. Apremium also has to be paid. These coverage limitationsor refusals have turned attention towards private funds,by way of captive insurance, for example. These serveabove all as deductibles, also called retention, andfacilitate access to reinsurance.

And that’s it! End of story! Nothing of the sort.Businessmen will always tell you that risks have to berun. Youngsters are told to knuckle down to theirstudies to be better prepared for the life hurdles up

Risk


always easy to lodge. When some form of moderationhas been arranged, the least moderator is forced to bowout if the case is unforeseen, something that oftenoccurs in our longwinded laws that try to fill everyloophole with blanket coverage and think they have goteverything covered. The most galling thing is that themore precise we try to be, the more interpretations wehave to trundle out and the more forks we have tonegotiate along the way.

The truth is that respect involves many morethings than a mere tilt against discrimination. There isno shortage of battlefields: witness the following. If Ihave some staff under me, respect for them should makeme want to save them from harm. Avoiding accidents,banning harassment, whether moral or otherwise,staving off illness, helping the family in case of death orthe employee himself in case of invalidity, guaranteeinga decent pension without frittering away the funds instockmarket speculation, all these could be said to befacets of respect for the work and the worker. If theemployer watches out for the long term, he or sheshould also guarantee the longest term possible in theemployment contract; no firm is viable in the long termwithout the input of its collaborators.

Running a socially responsible company istantamount to showing respect and concern for thecommunity we live and work in. Showing concern forpublic life, succouring wherever possible, abiding bythe laws are all only facets of what should or must bedone. We should not forget the ills of the present dayconcerning the climate and various forms ofpollution. This is not only a question of showingrespect for nature but just as much towards the humanbeings who suffer from it. It is also a good pathtowards respect for mankind as a whole; throughoutthe whole world vast swathes of people suffer fromhunger, from cold, from heat or drought, and all are inneed of aid. Even in the best-off countries, is it normalfor hardworking human beings to earn less moneythan they need for feeding and caring for themselvesor putting a roof over their head?

Much is said about the rights to be respected,acquired rights, common law but each termonly creates a new border; you are eitherinside or out. The most outrageous example is the set oflegal rules designed to weed out discrimination acrossthe board. Has that got anything to do with equality?; isthere really any chance of doing away with all forms ofharassment? The more rules we make the moreexclusions we create. In fact it suffices to flout any oneof the many key anti-discrimination nostrums to beexcluded ex officio. The more criteria we create thegreater the risk of falling short of at least one of them;those who enforce them do not usually have either theright or the power to waive them. And when will therebe any published statistics about those who areexcluded? Everyone, if he gives any thought to it,knows a victim of the procedure among his or hercoterie or neighbourhood, and the least appeal is not

Respect

subjective choice of the final decision taker. There aretwo common driving forces or, if you like, twocomplementary philosophies that should always be keptin mind.


Here we have a word that is almost taboonowadays, especially among the young.Almost nobody today wants to own up tohis or her mistakes because almost nobody wants toput them right. Insurers are probably to blame for ageneral misconception of this principle; when youhave caused some damage, never declare yourselfresponsible for it, otherwise you risk losing yourcoverage. The car accident report shows the way togo; acknowledge the facts and only the facts; for allyou know there might be rules that will eventually getyou off, and the professional insurer will make sureyou benefit from them.

Worse still; the various schemes of no-faultliability springing up like mushrooms cloud thejudgement of the youngest (and also the not soyoung). Even if you are not in fact responsible, act as ifyou are and compensate the «victims». It would havebeen simpler to invent some sort of expanded socialsecurity scheme covering all the ills we might sufferfrom, but this would soon have proven unfundable.The USA has taken things to extremes, allowinglawyers to wax fat on the most farfetched or tenuous

causes; if they miss their mark it will cost you little ornothing. But if they succeed the booty has to beshared with them, often on a fifty-fifty basis. How canwe then resist the lure of easy gains? Long forgotten,the responsibility of our grandparents is now almostmeaningless. All the professions, all the crafts arecontaminated; mistakes are no longer countenanced.Take doctors; they can never guarantee a resultbecause they are bound only by a best-endeavourobligation. Some of them can no longer securecoverage at an affordable price and therefore refuse tooffer certain care… because things might not alwayscome out as hoped and patients won’t put up with thisany more. Stories of this ilk could be told ad infinitumso let’s go back to the basic principle. Being

Responsibility


responsible is to take on a steady role; parents areresponsible for bringing up their children, somethingthat is not always evident in single-parent families.Politicians are (or should be) responsible for thecommon good. Manufacturers are responsible vis-à-vis their clients, but even so they have failed to takestrict product-safety measures. The host of serviceproviders is responsible for their quality and also theirprice. Where lies the problem? In the simple fact thateveryone is crystal clear about what responsibilitymeans in the case of others but much less sure when itconcerns themselves. There is therefore an urgentneed to get back to the simplicity or grandeur of thegolden rule «Do unto others as you would have themdo unto you»…

We hence come to a new formulation of RiskManagement. In fact it is only a melange of the threeRs. From the very beginning of life on earth there hasbeen risk, and there is no progress without it. Risk isstrange to us because it is nothing more thanuncertainty, with a positive or negative outcome thatcan never be guaranteed. We should get our headround this fact instead of always harping on aboutlosses and damage. For its part behaviour in regard torisk can be summed up in two words; respect andresponsibility. Respect is nothing more than payingbenevolent and prudent attention to fellow beings andnature. Responsibility is there only to link up withrisk, in assuming and enforcing the concomitantduties. We can come up with all the arrangements welike, invent all sorts of recipes, draw up complexmodels but if they do not tally with these twoqualities we fall into overkill. This overkill whittles riskdown into harmful circumstances, proving right allthose who see in risk only the negative results. Today’sfinancial catastrophe is a monstrous proof of this; wehave even forgotten to respect common sense and noone holds him or herself accountable for themanoeuvre any more. Do we need still morecatastrophes before getting back to a lucid view ofRisk Management? ❘


2009For the eighth straight year, FUNDACIÓN MAPFRE

presents its ranking of the top 25 insurance groups in

Latin America by premium volume, this time for 2009.

Three rankings have been compiled – Overall, Life and

Non-Life – and separate information is also included on

local and multi-national insurers.

CENTRO DE ESTUDIOS FUNDACIÓN MAPFRE

Latin America

of insurance groups in

RANKINGRANKING


After six years of economic growth, thecountries of Latin America and theCaribbean saw their GDP decline by 1.9percent in 2009. The impact of the international

financial crisis was acute in late 2008 and early

2009, although internal and external factors helped

speed up the process of recovery starting in the

third quarter of 2009: counter-cyclical measures

under taken by several countries of the region

through fiscal and monetary policy; the sustained

strength of some Asian economies, whose demand

for products from the region boosted the recovery

of exports; and the improvement in the U.S.

economy, which helped create a better scenario for

the economies of Mexico and Central America.

The insurance markets of Latin America

continued to show signs of strength, with nominal

increases in premium volume as measured in local

currencies in all countries except Chile. For yet

another year, the rise in the euro against some local

currencies did not favor converting premium

revenue into the single European currency. Indeed,

the Mexican peso's fall against the euro was one of

the factors that contributed most to the changes in

this ranking. On the other hand, the dollar's rise

against the euro had the opposite effect in other

markets, such as that of Puerto Rico, for instance.

Premium volume in Latin America totaled

75,769 million euros1 in 2009, with a nominal rise

of 10.7% compared to 2008. Events in the main

insurance markets of the region can be summed up

in the following way:

■ In Argentina, Non-Life insurance was once

again the main driving force in the market´s

development, with the biggest increases

coming in Automobile and Workmen’s

Compensation insurance.

■ In Brazil, life insurance known as Vida

PREMIUM VOLUME IN LATIN AMERICA TOTALED 75,769 MILLION EUROS IN 2009, WITH A

NOMINAL RISE OF 10.7% COMPARED TO 2008

1 Does not include Health insurance in Brazil, Life

Annuities and Retirement insurance in Argentina

or Pensions in Mexico.


RANKING OF INSURANCE GROUPS IN LATIN AMERICA IN 2009TOTAL

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

1

2

3

4

8

5

18

6

13

11

7

14

9

15

10

12

20

16

19

24

21

23

–

22

–

GROUPS

BRADESCO SEGUROS

ITAÚ/UNIBANCO HOLDING

MAPFRE

METLIFE

BANCO DO BRASIL

LIBERTY MUTUAL

SANTANDER

PORTO SEGURO

CNP

SUL AMÉRICA

G. NACIONAL PROVINCIAL

TRIPLE-S

AXA

ZURICH

BBVA

ALLIANZ

HSBC

AIG

GENERALI

INBURSA

MCS

MMM HEALTHCARE

MERCANTIL

SURAMÉRICA

LA PREVISORA

COUNTRY

BRAZIL

BRAZIL

SPAIN

UNITED STATES

BRAZIL

UNITED STATES

SPAIN

BRAZIL

FRANCE

BRAZIL

MEXICO

PUERTO RICO

FRANCE

SWITZERLAND

SPAIN

GERMANY

UNITED KINGDOM

UNITED STATES

ITALY

MEXICO

UNITED STATES

UNITED STATES

VENEZUELA

COLOMBIA

VENEZUELA

% ▲

12,1

-5,8

22,7

0,7

55,6

26,0

98,3

18,7

16,5

10,4

-8,9

14,8

-5,9

8,7

-7,2

-5,6

12,1

0,6

3,1

67,7

24,2

27,1

45,1

-0,7

19,3

PREMIUMS (millons €)

2008 2009

5.834

4.741

4.284

2.527

2.370

2.317

2.311

1.858

1.527

1.489

1.417

1.411

1.393

1.328

1.278

1.267

1.216

1.213

1.146

1.143

1.112

947

936

805

790

MARKET SHARE2009 (%)

RANKING2008

RANKING2009

7,7

6,3

5,7

3,3

3,1

3,1

3,1

2,5

2,0

2,0

1,9

1,9

1,8

1,8

1,7

1,7

1,6

1,6

1,5

1,5

1,5

1,3

1,2

1,1

1,0

Total of the top 10

Total of the top 25

Total for the sector

24.989

41.036

68.430

29.259

46.660

75.769

17,1

13,7

10,7

38,6

61,6

100

5.203

5.035

3.490

2.509

1.523

1.839

1.166

1.565

1.311

1.348

1.555

1.229

1.480

1.222

1.377

1.343

1.084

1.205

1.111

682

895

745

645

811

662

SUPLEM. INGLES:SUPLEM. INGLES, page 1 @ Preflight€¦ · Although a progressive change in the...

Documents

Transcript of SUPLEM. INGLES:SUPLEM. INGLES, page 1 @ Preflight€¦ · Although a progressive change in the...