Console Insurance Is A Ripoff

Back of envelope financial calculations: Warranties, fine; insurance, no!
statistics, decision-theory
2009-03-102012-02-02 finished certainty: highly likely importance: 4


Con­sider the poor con­sumer con­sid­er­ing ‘in­sur­ance’. In­sur­ance is offered for all sorts of things, and often the con­sumer buys it—even when he should­n’t. One of the prob­lems in an in­effi­cient mar­ket­place—­like the ones we often must pur­chase in­—is that there’s a of sorts in play: if the in­sur­ance was ‘fair’, the in­surer would make no profit, so why would they offer it at all? They’ll only offer one which makes them a profit. There­fore, all the in­sur­ances on offer are un­fair (y­ou’ll get less out of it than you paid) and you should­n’t buy any!

Of course, we know why one would pur­chase in­sur­ance: be­cause the risks one is in­sur­ing against are too large to be borne at any given time (even though one can pay for them even­tu­al­ly). A house burn­ing down, chemother­a­py, a car to­taled, etc. One buys in­sur­ance as a way to trade many small doable pay­ments for a sin­gle large im­pos­si­ble pay­ment. This is a valu­able ser­vice to you, so you don’t mind buy­ing ‘un­fair’ in­sur­ance; your lower is traded off against a smaller of your fu­ture ex­pens­es. (Peo­ple are well known to be ; the rich are less so than the poor, which is sad1.)

But not all in­sur­ance is of the ‘cat­a­strophic’ va­ri­ety. I’ve seen in­sur­ance offered on travel trips, air­plane flights, TVs, and even video game con­soles! And that in­sur­ance is ex­pen­sive, dozens or hun­dreds of dol­lars. It’s strange that peo­ple ap­par­ently think they can afford the steep in­sur­ance fees but not bear the cost of just buy­ing a new con­sole or what­ev­er. The irony is es­pe­cially rich when one con­sid­ers that peo­ple grossly over­es­ti­mate how un­happy they would be­come after ma­jor trau­mas, and also smaller losses2; how much less so if their iPod or Xbox broke? This is prob­a­bly due to the ; es­pe­cially ironic is that in­sur­ance—the op­tion to change one’s mind—­may sab­o­tage one’s en­joy­ment of the pur­chase3.

These pur­chases make no sense from the orig­i­nal ra­tio­nale for buy­ing in­sur­ance. Nor are these in­sur­ances triv­ial side-lines com­pa­nies run to hu­mor their con­sumers: they are pop­u­lar ser­vices (~31%4 of pur­chases in one sam­ple), and they are highly profitable5.

Worst Buy

Let’s look at a real ex­am­ple. The smaller the good, the more ob­vi­ously a bad deal these kinds of in­sur­ances are. I re­cently saw offers of con­sole in­sur­ance at (sand­wiched be­tween $100 ca­bles), and was struck at how ex­ploita­tive the offer must be. (Best Buy is no­to­ri­ous among geeks for its ex­tra­or­di­nar­ily high prices and poor cus­tomer ser­vice6.)

What a deal

The price varies per item, but let’s take a pre­sum­ably com­mon one. As of 2009-03-09, bestbuy.com offered an for $399.99. The ba­sic in­sur­ance plan on that Xbox 360 will run $59.99. That is, you will pay ~16% ex­tra for in­sur­ance (). And it’s good only for 2 years or so. What does this con­sid­er­able pre­mium buy you?

Our oblig­a­tions un­der this Plan will be ful­filled in their en­tirety if we re­place your pro­duct, is­sue you a voucher or gift card or re­im­burse you for re­place­ment of your prod­uct pur­suant to these terms and con­di­tions.

It buys you one re­place­ment; or:

After three qual­i­fied (3) ser­vice re­pairs have been com­pleted on an in­di­vid­ual prod­uct and that in­di­vid­ual prod­uct re­quires a fourth qual­i­fied (4th) re­pair, as de­ter­mined by us, we will re­place it with a prod­uct of com­pa­ra­ble per­for­mance of like kind and qual­ity not to ex­ceed the orig­i­nal pur­chase price.

3 re­pairs and a re­place­ment. And that’s it. But of course, it’s not un­con­di­tion­al. You can’t sim­ply walk into Best Buy after a few days and an­nounce you want a new Xbox 360. Here’s in­valid rea­sons to seek re­pair:

This Plan does not cov­er:

  1. dam­age to your prod­uct caused by ac­ci­dent (un­less you have pur­chased the op­tional ADH Cov­er­age), abuse, ne­glect, in­ten­tional phys­i­cal dam­age, mis­use (in­clud­ing faulty in­stal­la­tion, re­pair, or main­te­nance by any­one other than an au­tho­rized ser­vice provider), unau­tho­rized mod­i­fi­ca­tion, virus­es, ex­treme en­vi­ron­ment (in­clud­ing ex­treme tem­per­a­ture or hu­mid­i­ty), ex­ter­nal con­den­sa­tion, light­ning, sta­tic elec­tric­i­ty, fire, flood, in­sect in­fes­ta­tion, ro­dents, war, ter­ror­ism, com­puter soft­ware re­lated fail­ures (un­less you have the Vi-Spy Cov­er­age) or other ex­ter­nal caus­es;
  2. prod­ucts that have been lost or stolen. This Plan only cov­ers prod­ucts that are re­turned to us in their en­tire­ty;
  3. cos­metic dam­age to your prod­uct in­clud­ing but not lim­ited to scratch­es, dents and bro­ken plas­tic on parts, that does not oth­er­wise affect its func­tion­al­ity or ma­te­ri­ally im­pair your use;
  4. prod­ucts with a se­r­ial num­ber that has been al­tered, de­faced or re­moved;
  5. prob­lems caused by a de­vice that is not your pro­duct, in­clud­ing equip­ment pur­chased at the same time as your pro­duct;
  6. con­sum­able parts, such as bat­ter­ies, un­less ex­pressly pro­vided for here­in;
  7. dam­age to, or loss of any soft­ware or data re­sid­ing or recorded in your pro­duct…
  8. dam­age to your xBox 360 due to Mi­crosoft’s ‘

Well, shucks. That seems to cover just about every­thing that would dam­age my new Xbox 360! If it ships bro­ken—as is ex­tremely likely and seems to cover al­most all of the re­ported break­age rates7, then it’s the orig­i­nal man­u­fac­tur­er’s prob­lem (Mi­crosoft); if I break it by ac­ci­dent, it’s my prob­lem; if it gets scratched up, it’s my prob­lem (and god help me if the scratches are any­where near the se­r­ial num­ber! Then I can’t use the plan at al­l!); if my house dam­ages it, it’s my prob­lem; if one of the many me­chan­i­cal and elec­tronic prob­lems with the Xbox man­i­fests, it’s ei­ther my or Mi­crosoft’s prob­lem; etc. This does­n’t seem to cover very much, al­though I sup­pose it cov­ers if my Xbox stops work­ing for no rea­son what­so­ever one day. (Which is­n’t a ter­ri­bly com­mon prob­lem with video game con­soles.)

We’re in­ter­ested not in what hap­pens if we get lucky and im­me­di­ately the Xbox 360 breaks and the man­u­fac­tur­er’s war­ranty does­n’t cover it and Best Buy will give us a new one8—after all, if we only go by best case sce­nar­io, the lot­tery is the great­est deal around—but rather in what we can ex­pect on av­er­age; we have no a pri­ori rea­son to ex­pect to be luck­ier or un­luck­ier than av­er­age. So we use the ex­pected value for­mu­la:

Gen­er­al­ly, one only wants to en­gage in deals or ac­tions which have a pos­i­tive ex­pected val­ue. Does the Xbox in­sur­ance have a pos­i­tive val­ue? Let’s say I have no idea of the true rates9. Cer­tainly nei­ther you nor I have ac­cess to good in­for­ma­tion about how many of Best Buy’s cus­tomers ac­tu­ally use the in­sur­ance suc­cess­fully (although Best Buy surely does). But we can still put some bounds on it: we can ask the ques­tion, “how likely would Xbox-break­ing have to be be­fore the in­sur­ance has a pos­i­tive pay­off?” That is, we know the pay­off—it’s 340. And we know the ex­pected value we want by de­fi­n­i­tion is for it to be greater than 0; but we want to get more out than we put in, so we need to get back at least $60 if we want to just break-even. With 2 out of the 3 vari­ables solved, ba­sic al­ge­bra can now fig­ure out the fi­nal vari­able, the prob­a­bil­i­ty:

More details

So we need a 1-in-6 chance of re­deem­ing the in­sur­ance. But wait! The 1⁄6 does­n’t ap­ply di­rectly to the chance of break­ing an Xbox. Rather, it ap­plies to the con­junc­tion I ex­pressed be­fore: break­ing the Xbox and Mi­crosoft not cov­er­ing it and Best Buy mak­ing good10 and us ac­tu­ally us­ing it. As we all know, a con­junc­tion of a bunch of prob­a­bil­i­ties is go­ing to be less likely than any of the prob­a­bil­i­ties in them­selves. So the first prob­a­bil­i­ty, ‘break­ing the Xbox’ is go­ing to be more likely than just 1⁄6.

How much more like­ly? Well, we can play with some more num­bers here. Let’s see what hap­pens if we are very gen­er­ous & as­sign all the other prob­a­bil­i­ties some­thing like a 90% chance. We know the to­tal is 1⁄6, and we know the other 3 we’ve de­fined to 0.90, and we leave one vari­able to solve for, ‘break­ing the Xbox’ ():

So just by be­ing a lit­tle more re­al­is­tic, we see that for the in­sur­ance to make sense, we need an aw­fully high rate of Xbox break­age. And the nec­es­sary break­age rate for us to break-even be­comes still worse, still more in­cred­i­ble, if we were to be less gen­er­ous in the con­junc­tion.

I don’t know about you, but I don’t be­lieve I have a 1 in 4 chance of break­ing my new Xbox within the 2 years the in­sur­ance cov­ers. I be­lieve the chances are much lower11. And that makes this in­sur­ance one lousy deal.

Deeper down the (rabbit) hole

Nor is con­sole in­sur­ance the worst offer Best Buy has. It re­cently (Feb­ru­ary 2011) be­gan a cu­ri­ous buy/lease pro­gram12. In this scheme, the cus­tomer pays 10% up front, and in ex­change, if the cus­tomer re­turns the item for any rea­sons within 6 months, Best Buy will give them a 50% dis­count on a new/replacement item. It’s hard to see who this trade-in offer is for. One’s time is not free and one does­n’t want to switch prod­ucts too fre­quent­ly; and if we as­sume gen­er­ously that a com­pelling new prod­uct comes out ran­domly every 2 or 3 years, the num­bers are even worse.


  1. Chen et al 2009:

    The re­sults from the het­ero­gene­ity analy­sis are in­trigu­ing. In con­trast to prior em­pir­i­cal find­ings (Pad­man­ab­han 1995; Pad­man­ab­han and Rao 1993), we find that, as com­pared to high­-in­come con­sumers, low-in­come con­sumers are more likely to pur­chase ESCs. The analy­sis re­veals that they are likely to do so be­cause they are more sen­si­tive to the re­place­ment costs in the event of prod­uct fail­ure. Un­like in the case of au­to­mo­biles, where high­-in­come con­sumers buy ESCs to avoid main­te­nance be­cause they have higher time costs, in the elec­tron­ics prod­uct cat­e­go­ry, low-in­come con­sumers buy in­sur­ance to hedge against the out­-of-pocket costs of re­plac­ing the prod­uct. Ad­di­tion­al­ly, low­er-in­come con­sumers are also more pre­dis­posed to ex­er­cise the sav­ings ob­tained from pro­mo­tions to pur­chase ESCs.

    If the ESCs do, in fact, offer lit­tle val­ue, the re­sults im­ply a per­verse im­pact on con­sumer wel­fare. The lack of fi­nan­cial abil­ity of low-in­come con­sumers to re­place prod­ucts in­duces them to pay a po­ten­tially un­nec­es­sary and over­priced in­sur­ance pre­mi­um. High­-in­come con­sumers, for whom prod­uct re­place­ment is not a cause for anx­i­ety, in­cur a lower to­tal cost of prod­uct ac­qui­si­tion. These find­ings are some­what ironic in light of ob­ser­va­tions that poor sick pa­tients who are un­able to afford health in­sur­ance pay the high­est prices for drugs (e.g., Frank 2001). In the health do­main, poor pa­tients are un­able to afford in­sur­ance in a cat­e­gory where it is salu­bri­ous to buy it, but they opt for in­sur­ance in an area where the in­vest­ment is in­ex­pe­di­ent.

    ↩︎
  2. “If money does­n’t make you hap­py, then you prob­a­bly aren’t spend­ing it right”, Dunn et al 2011, Jour­nal of Con­sumer Psy­chol­ogy:

    Re­search on how well peo­ple cope with a wide va­ri­ety of trau­mas and tragedies-from heart at­tacks to ter­ror­ist at­tack­s-sug­gests that peo­ple are not the emo­tion­ally frag­ile crea­tures they often imag­ine them­selves to be (Bo­nan­no, 2004; Ubel, 2006).

    …re­search sug­gests that peo­ple don’t know much about their own psy­cho­log­i­cal im­mune sys­tems (Gilbert, Pinel, Wilson, Blum­berg, & Wheat­ley, 1998), and as a re­sult they over­es­ti­mate their vul­ner­a­bil­ity to neg­a­tive affec­t…re­search shows that this ex­pec­ta­tion is wrong. Ker­mer et al. (2006) gave par­tic­i­pants $5, and then flipped a coin. Par­tic­i­pants were told that if the coin came up one way they would get an ad­di­tional $5, and if it came up the other way they would lose $3 of their ini­tial en­dow­ment. Al­though par­tic­i­pants ex­pected to be more emo­tion­ally affected by the loss of $3 than by the gain of $5, they were not.

    …When pas­sen­gers on a train were asked to es­ti­mate how much re­gret they would feel have felt if they had missed the train by 5 min­utes or 1 min­ute, they es­ti­mated that they would have felt more re­gret in the lat­ter case than the for­mer. And yet, pas­sen­gers who had ac­tu­ally missed their trains by 1 and 5 min re­ported re­mark­ably lit­tle re­gret, and equally lit­tle re­gret re­gard­less of whether they had missed the train by 5 min or by 1 (Gilbert, Morewedge, Risen, and Wil­son 2004).

    For an­other good ex­am­ple of peo­ple fail­ing to un­der­stand them­selves, see the next foot­note on how re­turn poli­cies (and thus warranties/insurance?) can sab­o­tage one’s sat­is­fac­tion with a pur­chase.↩︎

  3. Dunn et al 2011:

    Un­for­tu­nate­ly, this handy men­tal mech­a­nism may ac­tu­ally be short­-cir­cuited by gen­er­ous re­turn poli­cies. offered par­tic­i­pants the choice be­tween prints of paint­ings by artists rang­ing from Van Gogh to El Gre­co. After par­tic­i­pants made their se­lec­tion, half of them were pre­sented with the equiv­a­lent of a gen­er­ous store re­turn pol­i­cy: they were told, “If you change your mind about which poster you want to take home be­fore you leave to­day or even any time in the next mon­th, you can just let me know and we will ex­change it for you.” The re­main­ing par­tic­i­pants were in­formed that no such ex­change would be pos­si­ble and that their choice was fi­nal. Par­tic­i­pants who knew they were stuck with the poster they had cho­sen re­sponded by in­flat­ing their ap­pre­ci­a­tion of it, see­ing the poster in a more pos­i­tive light than they had ini­tial­ly. In con­trast, par­tic­i­pants who knew they could ex­change their poster any­time were de­prived of this emo­tional ben­e­fit of com­mit­ment and found the poster no more at­trac­tive than they had be­fore se­lect­ing it (see also Frey, 1981; Frey, Kumpf, Ir­le, & Gniech, 1984; Gi­rard, 1968; Jeck­er, 1964). In­ter­est­ing­ly, how­ev­er, par­tic­i­pants failed to pre­dict this differ­ence and thought they would be equally happy whether they could ex­change their poster or not.

    ↩︎
  4. From “A Chance to Keep Up With New Tech­nol­ogy (for a Price)”, Stross 2011, New York Times:

    Pro­fes­sor Kalra was co-au­thor of a 2009 ar­ti­cle pub­lished in the Jour­nal of Con­sumer Re­search that ex­am­ined pur­chase records from the elec­tron­ics de­part­ment of an uniden­ti­fied re­tail­er. [Tao Chen, Ajay Kalra and Bao­hong Sun (2009), “Why do Con­sumers Buy Ex­tended Ser­vice Con­tracts”, Jour­nal of Con­sumer Re­search Vol­ume 36 (De­cem­ber), 611-623] An ex­tended ser­vice con­tract was bought in about 3 of every 10 trans­ac­tions.

    ↩︎
  5. From Stross 2011:

    Mr. Fassler says Best Buy has not dis­closed for many years how much ex­tended war­ranties con­tribute to its profits. Con­sumer elec­tron­ics re­tail­ing is “his­tor­i­cally a low-mar­gin busi­ness that is de­pen­dent on ex­tended war­ranties for profitabil­i­ty,” he says. “Per­haps ex­tended war­ranties have be­come even more valu­able to Best Buy re­cent­ly, as its con­tent busi­ness­es—­like movies and mu­sic—have shrunk.”

    War­ranty Week pro­vides ag­gre­gate and bro­ken-down fig­ures in “Ex­tended War­ranty Ad­min­is­tra­tors: While auto and PC man­u­fac­tur­ers have the top spots, in­sur­ance com­pa­nies and third party ad­min­is­tra­tors grab the bulk of the pie”:

    Just as in pre­vi­ous edi­tions we’ve sized the prod­uct war­ranty in­dus­try to be in the neigh­bor­hood of $25 bil­lion per year, in this edi­tion we’re sug­gest­ing that ex­tended war­ranties gen­er­ate in the vicin­ity of $15 bil­lion per year in pre­mi­ums paid. Not all that money goes to the ac­tual ad­min­is­tra­tors and un­der­writ­ers of the poli­cies, how­ev­er. In fact, we’re es­ti­mat­ing that roughly half is kept by the ac­tual sell­ers – the re­tail­ers and deal­ers in the world of war­ran­ty. Only $7.5 bil­lion passes through to the ad­min­is­tra­tors, we es­ti­mate.

    “The War­ranty Wind­fall” by R. Berner Busi­ness Week (2004), offers more de­tails on house­hold names:

    War­ranties cost vir­tu­ally noth­ing to mar­ket, and the prod­ucts they in­sure rarely need re­pairs. Says FTN Mid­west Se­cu­ri­ties Corp. an­a­lyst Daryl Boehringer: “It’s just pure profit flow­ing down to the bot­tom line.”

    Last year, profits from war­ranties ac­counted for all of Cir­cuit City’s op­er­at­ing in­come and al­most half of Best Buy’s, say an­a­lysts. They fig­ure that profit mar­gins on con­tracts are be­tween 50% and 60%. That’s nearly 18 times the mar­gin on the goods them­selves. For ex­am­ple, a four-year con­tract on a $3,000 flat-panel TV costs about $400. Best Buy gives its in­sur­ers $160 and keeps $240 for it­self.

    …As ser­vice con­tracts be­come more crit­i­cal to its bot­tom line, Best Buy has ac­tu­ally cut back on dis­clo­sure. The Rich­field (Min­n.)-based chain does­n’t re­port its war­ranty profits sep­a­rate­ly, though it used to give the per­cent­age of sales that the con­tracts com­prised. It stopped do­ing that after fis­cal 2001 and buried the num­ber in a rev­enue cat­e­gory la­beled “oth­er.” Then for fis­cal 2004 it stopped re­port­ing the “other” cat­e­gory al­to­geth­er.

    Cir­cuit City is more forth­com­ing. The Rich­mond (Va.)-based out­fit re­ports how much rev­enue the con­tracts gen­er­ate, along with the per­cent­age of sales they make up – but not the profit they pro­duce. For the year ended Feb. 29, it said its war­ranty rev­enue to­taled $326 mil­lion, or 3.3% of sales. Clear­ly, says SAFE’s Se­bas­tian, the re­tail­ers “don’t want to dis­close to J.Q. Pub­lic how much money they are mak­ing on these con­tracts.”

    Us­ing de­tails gleaned from in­dus­try sources, though, an­a­lyst Boehringer put to­gether es­ti­mates of just how lu­cra­tive these con­tracts are. For the year ended Feb. 28, he es­ti­mates con­tract profits ac­counted for 45%, or $600 mil­lion, of Best Buy’s $1.3 bil­lion op­er­at­ing profit. He fig­ures that with­out con­tract profits, Cir­cuit City would have posted an op­er­at­ing loss from con­tin­u­ing op­er­a­tions of $195 mil­lion last year in­stead of a $564,000 profit.

    ↩︎
  6. See eg. Wikipedi­a’s sec­tion, Google for “best buy sucks”, the break-down by cat­e­gory of com­plaints to the ; or you can just read For­tune Mag­a­zine’s “Best Buy’s gi­ant gam­ble” ar­ti­cle for your­self. (If you can read through this ar­ti­cle with­out re­al­iz­ing that Best Buy ad­mits that its cor­po­rate strat­egy is “charge suck­ers high prices”, then you need to work on your crit­i­cal think­ing skill­s.)↩︎

  7. Re­port­edly, Game In­former found a 54.2% break­age rate for Xbox 360s by sur­vey­ing read­ers; Ex­am­iner claims just <16.5% based on how many Xbox 360s are ac­tive on­line. Nei­ther fig­ure seems very re­li­able.↩︎

  8. A sit­u­a­tion in which the in­sur­ance is ob­vi­ously profitable, as what would have been an $800 ex­pen­di­ture be­comes a $460 ex­pen­di­ture—which saves a healthy $340.↩︎

  9. Chen et al 2009 note that Con­sumer Re­ports listed the over­all fail­ure rate of video game equip­ment at 9%.↩︎

  10. not a triv­ial con­sid­er­a­tion, given their cus­tomer ser­vice and the lengthy list of ex­cluded is­sues↩︎

  11. From Stross 2011 (NYT):

    Mark Kotk­in, di­rec­tor of sur­vey re­search at , says, “The sales­per­son often tells you, ‘This will give you peace of mind in case you need an ex­pen­sive re­pair. I would get it.’ But the odds of a prod­uct break­ing down dur­ing a typ­i­cal ex­tended war­ranty pe­riod are low.” Even if the prod­uct does break, the cost of re­pair is not much more, on av­er­age, than the cost of the war­ran­ty, he says. He con­cludes that ex­tended war­ranties “are a bad bet.”

    Ajay Kalra, a mar­ket­ing pro­fes­sor at Rice Uni­ver­si­ty, agrees. “All the sta­tis­tics are com­pelling: in al­most all cas­es, you should­n’t buy the ex­tended war­ran­ty,” he says.

    ↩︎
  12. Stross 2011.↩︎