The question of whether or not to buy warranties is one we all face, especially when buying big ticket electronic items. There is value in casting this question in probabilistic terms.

Firstly, we assume that repurchasing of the item will occur in the event that a warranty is not purchased and the item breaks. This is important because there is a distinct attraction to the idea of knowing that we will always get a repaired or replaced item. However, if we are going to repurchase the item this attraction exists regardless of whether one gets the warranty. The only question is going to be the price and the variance in the price.

As an aggregate, sellers of warranties profit off the sales. In many electronic stores, this is actually a large proportion of the profits. On average, the consumer is losing "expected value" by purchasing the warranty for it costs more on average than it is worth. There is actually a gap where it is possible that both parties win since the customer has to pay the companies profit on the repurchasing of a new item that the company doesn't pay if it replaces it - it does this at cost. If the average cost of the warranty is less than the average profit margin, it can be a positive expectation to buy the warranty. However, especially in electronics where item profit margins are thin and warranties are a high proportion of the cost, this is usually not the case and a reasonable working assumption is that the customer is experience a net average cost buying the warranty.

The advantage of getting the warranty is in minimizing the variance in expected results. If one purchases the warranty the expected costs are fixed at the price of the item plus the warranty. If we don't buy the warranty there is a variance in the outcomes costing either the price or possibly twice the price of the item. By buying the warranty we are exchanging an average loss in money for the certainty in price. Note that we are not changing the certainty of always having the item - by our assumption we are repurchasing it and so will have one regardless.

The question of whether to purchase a warranty then comes down to ones subjective valuations of how worthwhile this price certainty is. As a general guideline, if one has a reasonable amount of money such that repurchasing in the future would be possible then not buying the warranty is saving money and the variance reduction is not that important. For someone without a lot of money who perhaps saves for a long time to buy the product, reducing the variance may be more important than maximizing profit since a second purchase of the product would not necessarily be able to occur or would be especially difficult. Ironically, it is thus those with less money who should be more inclined to take the more expensive option as they value the variance reduction more.

Finally, it should be pointed out that there is a distinct information asymmetry involved. It is very difficult for a consumer to know enough about the products to do the relevant actuarial computations to see how likely a product is to break and thus how much the warranty is worth. We have no good heuristic for determining this. The companies however study this extensively and so you can be certain that it profits them while having little idea if it profits you. I am thus pretty skeptical about the validity of the entire business.

Firstly, we assume that repurchasing of the item will occur in the event that a warranty is not purchased and the item breaks. This is important because there is a distinct attraction to the idea of knowing that we will always get a repaired or replaced item. However, if we are going to repurchase the item this attraction exists regardless of whether one gets the warranty. The only question is going to be the price and the variance in the price.

As an aggregate, sellers of warranties profit off the sales. In many electronic stores, this is actually a large proportion of the profits. On average, the consumer is losing "expected value" by purchasing the warranty for it costs more on average than it is worth. There is actually a gap where it is possible that both parties win since the customer has to pay the companies profit on the repurchasing of a new item that the company doesn't pay if it replaces it - it does this at cost. If the average cost of the warranty is less than the average profit margin, it can be a positive expectation to buy the warranty. However, especially in electronics where item profit margins are thin and warranties are a high proportion of the cost, this is usually not the case and a reasonable working assumption is that the customer is experience a net average cost buying the warranty.

The advantage of getting the warranty is in minimizing the variance in expected results. If one purchases the warranty the expected costs are fixed at the price of the item plus the warranty. If we don't buy the warranty there is a variance in the outcomes costing either the price or possibly twice the price of the item. By buying the warranty we are exchanging an average loss in money for the certainty in price. Note that we are not changing the certainty of always having the item - by our assumption we are repurchasing it and so will have one regardless.

The question of whether to purchase a warranty then comes down to ones subjective valuations of how worthwhile this price certainty is. As a general guideline, if one has a reasonable amount of money such that repurchasing in the future would be possible then not buying the warranty is saving money and the variance reduction is not that important. For someone without a lot of money who perhaps saves for a long time to buy the product, reducing the variance may be more important than maximizing profit since a second purchase of the product would not necessarily be able to occur or would be especially difficult. Ironically, it is thus those with less money who should be more inclined to take the more expensive option as they value the variance reduction more.

Finally, it should be pointed out that there is a distinct information asymmetry involved. It is very difficult for a consumer to know enough about the products to do the relevant actuarial computations to see how likely a product is to break and thus how much the warranty is worth. We have no good heuristic for determining this. The companies however study this extensively and so you can be certain that it profits them while having little idea if it profits you. I am thus pretty skeptical about the validity of the entire business.

**Thoughts on this post? Comment below!**

**Share this post: **

## Post a Comment