قالب وردپرس درنا توس
Home https://server7.kproxy.com/servlet/redirect.srv/sruj/smyrwpoii/p2/ Business https://server7.kproxy.com/servlet/redirect.srv/sruj/smyrwpoii/p2/ Tesla Insurance: Buffett has a mistake

Tesla Insurance: Buffett has a mistake



Warren Buffett, Chairman and CEO of Berkshire Hathaway (BRK.A) (BRK.B), of which GEICO is a wholly-owned subsidiary, recently announced in response to Tesla's announcement of the planned offering of insurance : "The success of the car companies entering the insurance business is probably as likely as the success of the insurance companies entering the automotive business."

My opinion is that this statement is incorrect, as I will explain below , first by showing how the risk calculation of traditional carriers is wrong, and then explains how vendor risk models are likely to work better and finally demonstrate how this can change the whole industry.

Explaining the argument "This is too expensive or difficult" Many people, when they hear that car makers are planning to get insured, instinctively react with a knee reaction that this is impossible. Insurance carriers have spent billions on infrastructure, regulatory approvals and actuarial risk assessments, and have to keep cash to cover their claims. That's all the money Tesla does not have.

This is an incorrect conviction. There is a concept for MGA (or MGU), which is a type of agency that has the right to be branded as a carrier using a "forward carrier" relationship that is an existing insurer at the end of a partnership. MGA deals with pricing and emissions (often with the carrier), marketing, customer acquisition and policy service, the leading carrier deals with receivables and money management and shares profits from the joint venture.

Many new carriers start out as MGAs. Two remarkable examples are:

Insurers are likely to follow the same pattern. In the case of Tesla, they have already made a partnership with Markel, a reinsurer, seemingly a leading carrier. They also stated the need to buy a shell, possibly an existing MGA, which is already licensed in many countries. All that would be left to them is to build software to attract customers and service the policy. This should be a cake for Tesla, which already has a huge team of developers and a strong online presence.

Now that the "impossible" myth is defeated, let's look at how car manufacturers are ready to go beyond current insurers. in their own game

How traditional risk calculations are wrong

Ratemaking for auto insurance traditionally uses rating tables to calculate a given risk for the insured against the underlying risk. Base risk is a randomly selected profile defined by a carrier, typically to represent their typical exposure. For example, the main venturer's profile may look like this:

  • 45-year-old married man, no children at driving age
  • Bachelor's degree, software developer, owns a home in an urban environment
  • 10,000 miles per year for work
  • No tickets or claims over the past 5 years

This primary risk profile has been evaluated for base rate risk. Then each variable of the risk profile is expanded into a rating table listing each potential value or set of values ​​and determining a modifier or multiplier for the relativity based on the carrier's determination of how each value affects the risk

A rating table works, part of a gender / gender collision scorecode may look like this:

Age Men's Modulator COLL Female COLL Modifier
16-18 1.50 1.22
19-21 1.35 [19659020] 1.19 1.25 1.25 1.25 1.25 1.25 40 1.05 0.94
41-45 1.0 0.9

And so on …

to receive bonuses for each type of cover (collision, in-kind, property damage, bodily injury, etc.). Then the coverage level premiums are collected to reach a total premium, for example:

Item Cover Premium
BI 100/300 $ 100
500 $ 200 $ 200 $ 500 $ 200 There are many factors that include a person's risk profile, including:

$ 196

Age and Sex

  • ] Occupation
  • Level of education (where permitted by law)
  • Credit rating (where permitted by law)
  • Assess and safety of the vehicle
  • Use of the vehicle (work, leisure time, riding sharing, etc.)
  • Annually propelled vehicle
  • Records (Tickets, Accidents, etc.)
  • previous receivables
  • Garaging Address
  • Etc.
  • Each of the possible values ​​of these factors is listed in a sim table Similar to the one presented above, the modifiers for each value correspond to the carrier's risk assessment for that value

    This tabular valuation method has been used more than 100 years.

    The inherent disadvantage of this type is the perspective and evaluated and not for individuals . For example, in the segment of 26-30 year old men, in my sample table above, there is a 15% assumed increased collision risk compared to Base Risk due to the relative number of driving claims or abnormalities that the carrier has adopted at this age and gender. However, this does not take into account the fact that a part of the population of 26-30 / m leads to a higher than average risk for the group, which leads to an increase in the average risk for the whole group. Some men aged between 26 and 30 are extremely safe drivers and pose a lower risk than others who are extremely dangerous drivers. While other factors are taken into account in this exercise (for example, ticket history may be used to distinguish safer individuals from this category), it is not yet considered as a guide to each individual driver due to the low fidelity or the data holes, especially with regard to the driver's daily behavior. Therefore, in the traditional valuation model, for each carrier, some drivers will pay more expensive, while others will pay less for their actual risk.

    When this becomes a problem for carriers, this means [2] carriers 2 calculate percentages in the same way . For various reasons, each carrier has developed its own speed modifiers for each value of the variable. Thus, Carrier A may punish an age group more strongly, while Carrier B may place more weight on ticket history in risk assessment. That is why a man with 26-30 years without tickets can get a better price with Carrier B, while one with tickets will stick to Carrier A.

    Finally, the ability of drivers to self-select The most cost-effective pricing model for the carrier is inefficient for carriers – each of them will eventually threaten a particular population of under-valued risk, resulting in higher average loss rates than necessary. [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] [2] the world of auto insurance rating. By plugging a device into the car's OBD port or installing a smartphone application, drivers allow their insurance operators access to information about their actual driving behavior, allowing carriers to assess the risk more accurately for each individual driver. Measured behaviors may include:

    • High acceleration and heavy delays
    • How long does it take to drive
    • What hours of the day are spent driving
    • (For phone applications) Time spent on the phone during driving [19659088] This is an improvement on the traditional model of rating as it offers carriers a view of actual driving behavior, where in the past it could only be inferred from other data such as age, gender and vehicle records.

      Even telematics systems have some drawbacks:

      1. Most telematics systems are only consumed for a short period of time (eg 3-6 months), during which drivers are prompted to change their driving behavior, 19659012] In this time period, limited actual driving data is collected. Acceleration and deception only draw a small portion of the picture of how risky driver behavior is.
      2. Once the use of telematics is discontinued, no change in driving behavior (good or bad) is measured. Since carrier telematics is a switching system that is used by only a part of customers, carriers should stimulate customers by offering discounts for positive results without penalizing negative results.

        Why manufacturers' telematics is better

        OEMs come equipped with very powerful, connected computer systems that have peripheral input devices capable of recording other vehicles and road conditions around the car, GPS and mapping systems capable of speed limits, steering wheel sensors and pedals, and more. , Tesla (TSLA) is the leader of this front, with its Autopilot system using cameras, radar and ultrasonic sensors, as well as advanced processing technology to analyze and identify the risks that even people often miss. dozens of additional data points above what the traditional telematics device is capable of. They can also assemble actual driver behavior.

        A few years ago I created a prototype of the main telematics application using GM developer tools, and I was shocked by how much real-time driving data is available at all times

        With a flow-processing solution, cars will have the opportunity to analyze very precise driving behavior, such as:

        1. How often (and how much) does the driver speed relative to actual speed limits?
        2. How often does the driver follow too close or the back door?
        3. How aggressively the driver changes (acceleration and steering speed)?
        4. How aggressive does the driver get on the pedals (accelerator and pedal angular speed)
        5. Does the driver often change bands with other cars in the dead zone?
        6. Does the driver often work with red lights or roll over the stop signs?

        And so on …

        More than all the risk data currently analyzed in traditional assessment, these specific driving behaviors give a much more specific insight into how each real driver manages his car and how this behavior can to increase or reduce the risk of

        Hence, it follows that the car manufacturers' risk models will ultimately exceed the performance of traditional carriers for risk assessment for each particular individual.

        Why current insurers can not do too much

        Logical progression would make people believe that traditional operators will follow original equipment manufacturers by collecting telematics from on-board systems. However, existing operators will have to ask drivers to install applications on their cars, adding friction.

        In addition, historic driving data will not be immediately available to traditional operators, so the tariffs will need to be adjusted later on after the data is collected. Without the retrospective analysis data, the original quoted premiums of existing operators can not be as attractive as possible at the time of purchase.

        However, car manufacturers – especially those who build FSD systems – may already have historical data and will simply be able to request permission to access pre-anonymous driver recordings in the citation process. In the case of Tesla, drivers have already chosen to manage data collection to help with AutoPilot training. The same data can be used to calculate the risk and generate a real-time user rating

        Other Benefits to OEM Insurers

        Producers will have some additional benefits on their part, helping them compete with traditional operators in other areas:

        1. They will have more accurate data on repair costs as they will focus on a car they produce themselves.
        2. They will be able to offer replacement OEM parts for less as they will profit from buying part of OEM at the back end. Most insurers offer only parts for spare parts that consumers perceive as substandard.
        3. They will potentially be able to direct customers to their own network of discounted service centers, reaping some benefit.
        4. They will be able to directly offer existing owners and buyers of new cars at the time of purchase, with a fraction of the price of traditional operators (which reportedly pay $ 100- 500 CAC).
        5. The quote process will be significantly simpler, as VET, at least in the Tesla case, already has the bulk of the necessary data from its customers.

        the party's experience should be expected to be better than the OEM insurer.

        How do I predict everything that affects the insurance industry

        So if manufacturers can evaluate individuals better? The insurance works because the risk is measured collectively as a pool, so the total value of claims for the whole will remain the same, right?

        Well, yes … WHY …

        When manufacturers can very accurately identify drivers who are anomalously low risk (eg a 20-year-old man who drives very safely) and will be able to provide those individuals with lower rates to meet their actual individual risk, these anomaly low-risk drivers are likely to massively migrate to manufacturers for cheaper, better insurance.

        With a risky population that sticks to traditional carriers, their loss rates will increase and force them to raise rates. As a result, this will probably push the average risk level and manufacturers' insurance offers, ultimately leaving traditional operators only with the highest risky part of the population.

        With loss ratios for OEM insured vehicles that go out of control, some major operators may refuse to insure whole car brands, knowing the risk is too heavy. The rest will migrate to non-standard carriers that will benefit from the high-risk population (their specialty).

        Any light at the end of the tunnel?

        Unless traditional insurers find a way to collect and absorb more accurate telematic data in their risk models, I do not see how they can compete with future generations of OEM insurers. This will be difficult because manufacturers can collect this data internally, while approved users will have to install driver applications to get the same data (again, the worst drivers probably will not be included.)

        I see only two ways in which traditional operators can compete with OEM insurers:

        1. If they can convince the policy, the manufacturers or carmakers that drive should be openly shared between manufacturers and insurers, or
        2. if they ematichnoto third-party app received critical mass, thus collect comprehensive data driver that can be sold to insurers for assessment.

        I do not see any of them Scenarios as a short-term probability. While we may see this in the future, I would expect in the next few years that current operators will lose their best customers to VET insurance offers and will end up trying to get them back. So, I believe that Warren Buffett is wrong and actually has it back when it assumes that car manufacturers can not be successful in insurance. I believe that the original equipment manufacturers actually have a lead. And if they do not move on to the ASAP telematics strategy, they are the ones who will struggle to survive.

        Disclosure: I am / we are a long TSLA. Написах тази статия сама и тя изразява моята собствено мнение. Не получавам обезщетение за него (освен от Seeking Alpha). Нямам бизнес отношения с нито една компания, чиито акции са споменати в тази статия.

        Допълнително разкриване: ОПОВЕСТЯВАНЕ: НЯМА ФИНАНСОВ АНАЛИЗ ИЛИ СЪВЕТНИК. ВСИЧКО СЪДЪРЖАНИЕ, КОЕТО ДОПЪЛВА ДА СЕ ПОТРЕБА АЛФА Е САМО МНЕНИЕ, И НЕ СЪСТАВЯ СЪВЕТИ ЗА КУПУВАНЕ ИЛИ ПРОДАВАНЕ НА ЛИЦА. МОЛЯ, ИЗПОЛЗВАЙТЕ СВОЯ СОБСТВЕНО ИЗСЛЕДВАНЕ И КОНСУЛТИРАНЕ НА ФИНАНСОВА ПРОФЕСИОНАЛНА ПРЕДИ ИНВЕСТИЦИЯТА. : "редовен", "домакин": "https://seekingalpha.com", "emailHost": "https://email.seekingalpha.com", "про": невярно, "proPlus": невярно, "contributorCenter" : фалшива "realHostName": "https://seekingalpha.com", "isCms": фалшиви "cancelPV": фалшиви "isSharkPreview": фалшива "usersOnSite": "7568637", "assetHosts": [“https://static.seekingalpha.com”,”https://static1.seekingalpha.com”,”https://static2.seekingalpha.com”,”https://static3.seekingalpha.com”] "moneData": { "PARAMS": "new_navigation_ab = nav_white_right"} "assetHost": "https://static.seekingalpha.com", "userEchoHost": "https://feedback.seekingalpha.com", " ENV ": {" отклонение ": фалшива" спиране ": фалшива," производство ": вярно," тест ": фалшива}" gaAccountId ":" UA-142576245-1 "," gaLegacyAccountId ":" UA-1466493- 1 "," comscoreAccountId ": 8500672," fbAppId ":" 624608951014846 "," twitterAccountName ":" SeekingAlpha "," rollbarToken ":" 5edf110be2fc4cecb32637fc421111e2 "," perimeterXAppId ":" PXxgCxM9By "," embedlyKey ":" a6da93fdfc49472 099ce63260954716b "," т.т. ": фалшива" чат ": {" хо ва ":" https://rc.seekingalpha.com "}}" pageConfig ": {" опресняване ": {" активен ": фалшива}," данни ": {" статия ": {" ID ": 4274506, "Заглавие": "Застраховка на Тесла: Бъфет има грешка", "заглушаване": false, "primaryTicker": "", "prioritizedTicker": "gm", "primaryIsCrypto": null, "isTranscript": false, "isSlides" : false, "twitContent": "Застраховка на Тесла: Бъфет има грешка https://seekingalpha.com/article/4274506-tesla-insurance-buffett-got-wrong?source=tweet","isProArticle":false,"isProPaywall ": фалшива" paywallReason ": нула" isArchived ": фалшива" inEmbargo ": фалшива" isAuthorNewsletter ": фалшива" titleTest ": нула" archiveOn ": 0.0," isProNoEmbargo ": фалшива" URL ": "https://seekingalpha.com/article/4274506-tesla-insurance-buffett-got-wrong","i sFreeMpArticle":false,"isFreeProArticle":null,"isInsight":false,"insightSlug": "", " price_at_publication ": нула" date_at_publication ": нула" closest_trading_day ":" 2019-07-11 "," теми ": [“long-ideas”,”contrarian”,”sa-exclusive”,”us”,”article”]," from_liftigniter ": фалшива" isAnyProArticle ": фалшива" allowMpPromotion ": фалшива" article_datetime ":" 2019-07-11T11: 59: 05.000-04: 0 0 "," date_epoch ":" 1562817600 "," isEtf ": false," taggedUrlsHtml ":" u003cspan itemscope itemtype = 'http: //data-vocabulary.org/Breadcrumb' t /seekingalpha.com/analysis/investing-ideas/all 'sasource =' article_navigation 'itemprop =' заглавието 'u003eИнвестиращи идеи u003c / span u003e u003c / a003e u003ca href = 'https: //seekingalpha.com/analysis/investing-ideas/long-ideas' sasource = 'u003ca href =' https: //seekingalpha.com/analysis/investing-ideas/long-ideas 'sasource =' u003span itemprop = 'заглавие' u003eLong Идеи u003c / span u003e u003c / a003e u003c / span u003e "," isFidelityEducationPage ": false," contentData ": null "marketingBullet": нула "showPastPodcast": фалшиви "publishDate": "07.11.2019"} ", автор": { "ID": 104,235, "потребителско име": 4512561, "заготовка": "Steven- moseley "," exclusiveResearch ": null," tagId ": 578977," име ":" Steven Moseley "," picture_url ":" https://static1.seekingalpha.com/images/users_profile/004/512/561/big_pic .png "," is_br and_author ": false," show_managed_account ": false}," коментари ": {" discussion_status ": 0," discussion_message ":" Коментарите са забранени за тази статия "}," brand ": null," firstResearchAuthor ": {}," тип страница ":" статия "," articleModeratedMsg ": нула" indexTickers ": {}" authorSentiment ": [{“primary_ticker”:”brk.a”,”type_id”:2},{“primary_ticker”:”brk.b”,”type_id”:2},{“primary_ticker”:”tsla”,”type_id”:1}]}," реклами ": {" карета ": [{“container”:”article-left-slot-2″,”data”:{“name”:”/6001/sek.analysis/stock-ideas/long-ideas”,”disable_collapse_empty_div”:false,”targeting”:{“tile”:160},”size”:[[160,600]," течност "]", ул " : "160×600, те ност"}}, { "контейнер": "статия ляв слот-3", "данни": { "име": "/ 6001 / sek.analysis / борсови идеи / дълги идеи", "disable_collapse_empty_div": фалшива "насочване": { "плочки": 162}, "размер": [[160,600]"течност"] ", ул": "160×600, течност"} "забавяне": вярно, "whenOutOfView ":" статия ляв слот-2 "," лепкава ": вярно}, {" контейнер ":" статия-десния слот-1 "," данни ": {" име ":" / 6001 / sek.analysis / борсови идеи / дълго-идеи "," disable_collapse_empty_div ": фалшиви" насочване ": {" плочки ": 1}," размер ": [[300,250][300,600]," течност "]", ул ":" 300×250,300×600, течност "}" CLS ":" mb25 "," огъване ": вярно}, {" контейнер ":" статия-десния слот-2 "," данни ": {" име ":" / 6001 /sek.analysis/stock-ideas/long-idea s","disable_collapse_empty_div":false,"targeting":{"tile ": 100}," размер ": [[300,100]]" ул ":" 300×100 "}" CLS ":" mb25 "}, {" контейнер ":" статия-десния слот-3 "," данни " : { "име": "/ 6001 / sek.analysis / борсови идеи / дълго-идеи", "disable_collapse_empty_div": фалшиви "насочване": { "плочки": 2}, "размер": [[300,252]" течност "]", ул ":" 300×252, течност "}" CLS ":" mb25 "," естествени ": истински}, {" контейнер ":" статия-средна слот-1 "," данни ": { "име": "/ 6001 / sek.analysis / борсови идеи / дълго-идеи", "disable_collapse_empty_div": фалшиви "насочване": {} "размер": [[640,40]] "ул": "640×40" }}, { "контейнер": "ad_728x90", "данни": { "име": "/ 6001 / sek.analysis / борсови идеи / дългосрочни идеи", "disable_collapse_empty_div": ф алшиви, "насочени": { " плочки ": 728}," размер ": [[728,90]]" ул ":" 728×90 "}" забавяне ": вярно," б ": 500}, {" контейнер ":" instream_recommendation_ad "," данни ": { "име": "/ sek.analysis / борсови идеи / дълго-идеи / 6001", "disable_collapse_empty_div": фалшиви "насочване": { "плочки": 50}, "размер": [“fluid”]"ул ":" течност "}" забавяне ": вярно," б ": 500}]" testScroll ": вярно," увреждане ": фалшива" КВС ": {" г ": [“analysis”,”investing-ideas”]," т ": [“stock-ideas”,”long-ideas”,”contrarian”,”sa-exclusive”,”us”,”article”]"помощ": "4274506", "с": "Стивън-Mosel ей "," CNT ": [“4″,”8″,”oil”,”fed”,”taxes”,”bny”,”fccs”,”ssga2″,”ssga6″,”191″,”fnk1″,”fnk2″,”port”,”fnk3″,”fnk4″,”loan”,”NYL1″]," е ": [“gm”,”brk.a”,”brk.b”,”tsla”]," ab_new_navigation_ab ":" nav_white_right "}}," пътища ": {" Int ": {" adsAPI ": {" SRC ":" HTTPS : //static.seekingalpha.com/assets/api/ads-ee51eaa4e0c3d03a6b84fbe3ea865d7ac849dc5c2a21a4b3ac2640bde0bc15ae.js "," ID ":" SA-реклами API "}}," разширени ": [{“src”:”https://js-sec.indexww.com/ht/p/183642-162053219843577.js”,”id”:”ie”},{“src”:”https://www.googletagservices.com/tag/js/gpt.js”,”id”:”gpt”},{“src”:”https://pub.doubleverify.com/pub.js?ctx=818052u0026cmp=DV157324″,”id”:”doubleverify”},{“src”:”https://sb.scorecardresearch.com/beacon.js”,”id”:”sb”},{“src”:”https://ssl.google-analytics.com/ga.js”,”id”:”ga”},{“src”:”https://connect.facebook.net/en_US/sdk.js”,”id”:”facebook-jssdk”}]} ," lastRequested ":" 2019- 07-11 13:08:04 -0400 "," SlugsPrices ": {" disabled ": false}," proOpenHouse ": {" active ": false," end_date ":" 2017-03-20T06: 00: 00.000- 04:00 "}" proFlashSale ": {" активен ": фалшива" крайна_дата ":" 2017-03-23T00: 00: 00.000-04: 00 "}" saSource ": нула," име ":" статия "," useSQuoteBackup ": нула" inFreeArticleTest ": фалшива" featureCampaignVersion ": нула" featureCampaignFlow ": нула," skipAdoricFeatures ": фалшиви}" headerConfig ": {" noNotificationsMenu ": нула" nonFixed ": нула, "tabless": нула "activeTab": "борсови идеи"} "модули": {} ", изисква": []"trackq": []"изключения": []"gptInit ": фалшива}
        Source link