Hacker News Reader

Comment by transfire | Hacker News Reader

This is really the fundamental problem of Capitalism which we are now regularly butting are heads against. If it were possible, say, to create a light bulb that lasted 1,000 years for a dollar. Who could afford to make such a thing? After the initial flurry of sales, demand would drop off precipitously. And as the world approaches peek population, there would be fewer and fewer sales. And that's just the logistical problem. This story speaks to the further moral dimension in which it is always more profitable to corner a market and force on consumers ever increasing reoccurring cost.

View on HN

If it were possible, say, to create a light bulb that lasted 1,000 years for a dollar. Who could afford to make such a thing?

Anybody who wants to get rich.

Let's say that you can sell such a light bulb, and that it's so great that everyone in the world buys one, on average. And say you can sell it for a dollar, but you only make a cent of profit on each one. Why, that'd only leave you with. . . $70,000,000 dollars.

Well that's not too bad, but really not great either. It'd certainly let you retire to a tropical island for the rest of your life. But you could do better. If the light bulb is so great, wouldn't everyone be willing to spend more than a dollar on one? After all, those inferior light bulbs cost about twice as much, so if your bulbs are better than others should be willing to pay at least that much. Let's knock the price up to $2, then. That puts the profit per bulb at $1.01. So then you can simultaneously destroy the light bulb industry (fun!) and put a cool 7 billion dollars in your pocket (profit!).

Who would choose not do do such a thing? Certainly most anybody who's motivated by profit (like, say, a capitalist) would jump at the opportunity.

If it took 1 person to make, sell, and distribute those things, then $70 million are enough to retire comfortably. If it takes 100 people, then you could still retire, if you lived somewhere cheap.

More realistically, you might sell a few hundred thousand to the developed world, and then sales fall off a cliff, because it takes a lot of work to distribute outside your home market. Total payoff: Less than 2 months' salary.

That's the concern with the new lights. The industry is built on 1,000-hour bulbs and regular replacements. We currently have a boom as incandescents get replaced by fluorescents and fluorescents get replaced by LEDs (my father hates fluorescent light). Once that is done, who is going to buy bulbs at the rate that the factories can produce them?

This thread has gotten way out of hand with the thought experiments that reveal more about the author than the subject... but I can't resist!

If it took 100 people to make/sell/distribute, and you are the owner of the company, you're probably going to still be rich. Figure 99 employees at $150k fully loaded annually would cost you 15 million dollars, leaving 55 million to retire on.

Of course, if you are an employee, you would only have a job for 1 year.

Even assuming the hypothetical price does not refer to just production costs per bulb, but also initial capital outlays for a manufacturing plant etc., you're still assuming there are no marketing and sales costs, and no distribution costs, and no costs to cover returns (want to claim it lasts a 1,000 years? better be prepared to replace each and every bulb that fails for some reason or face lawsuits).

You'd also need capital to stand up to much larger manufacturers who'd go for an all out assault to find real or imagined flaws with your product (it'll destroy the environment - think of the dying baby seals!).

Some products costs many times their manufacturing cost to sell and distribute.

> After all, those inferior light bulbs cost about twice as much

That may sound like a good thing for you, but it's not a given. You've just created a reason for people to go "hold on, if this bulb lasts so long, why does it cost so little? what's wrong with it?" (at which point the less ethical of your competitors will whisper in their ears: "think of the dying baby seals; we can't prove that's what they make their filament of, but ...")

You'll need to convince people that you're selling them a genuine product, because it sounds too good to be true and a lot of people would assume it is too good to be true. Convincing them would cost a lot of money.

The irony is that you might be better off promising a much shorter timespan (you'd still need a fortune for marketing campaigns) and/or hiking the price up.

But if you're promising a shorter lifespan, why not cut manufacturing costs and achieve a shorter lifespan and increase your profit? It's what your competitors will do.

And of course you'd face clones of your products - patents or no patents there'd be people trying to find workarounds, or just blatantly ignoring any patents and hoping to make a profit before you can stop them (or hope they can make it too expensive for you to pursue)

> Who would choose not do do such a thing? Certainly most anybody who's motivated by profit (like, say, a capitalist) would jump at the opportunity.

The article points out that a cartel of the biggest manufacturers at the time saw things opposite: They saw increasing how long they'd last as a threat to the extent that they went to extreme lengths to prevent it from coming to pass.

I think that's a pretty good indication that "most anybody who's motivated by profit" would not necessarily jump at the opportunity to create a longer life product.

You kind of ignoring that we live in the age of the internet and word-of-mouth can work a lot better than sales/marketing. Returns too, only matter if it actually does fail, which given the premiss it doesnt. No reason why a exponential improvement in lightbulbs would be less popular than anything else people have looked at and said this is way better than that.

No, I'm not ignoring we live in the age of the internet. The age of the internet has not stopped spending on marketing at all, it's just changed where it gets deployed.

Returns matter because it will fail no matter what, given that there will be a non-zero change of failure in every step from manufacture to end user distribution, to failures directly caused by the end user (what do you mean it it's my fault it broke when I tried to screw it into the wrong type of fixture and used lots of force?) to forces of nature (lightning frying everything). End users will try it whether they're right, or whether they they hit the thing with a hammer.

If the premise is "is indestructable", then we are in fairy tale land, well past "a bit far fetched hypothetical land". My expectation was that the premise was more akin to "will last 1000 years absent production failures and when treated/used according to spec".

I've dealt with customers demanding money back on an internet subscription because they didn't realise they needed a computer (this was before smart phones and the like) to use the internet.You will need to deal with returns.

Even if only rebuff attempt at claiming money back.

> No reason why a exponential improvement in lightbulbs would be less popular than anything else people have looked at and said this is way better than that.

Name one single such product that managed to get worldwide penetration with no marketing from anyone.

The point is not that such a bulb would not become popular if it came about, but that the cost of making any physical product a world wide success are high enough that a cheap product that's effectively destroying it's own market is not likely to be a good investment even if the potential final "one time" sales could be fairly large.

Consider that e.g. OSRAM when they spend marketing money is maintaining a market that keeps buying product. Every customer they gain is likely to buy bulbs multiple times, so their marketing cost per acquired customer is likely to be far lower than an unknown entrant who needs to keep selling to new people.

Keep also in mind that even with a promise of a 1000 year lifetime, it will take many years before the product will have proven itself sufficiently that people will have first hand evidence that it even outlasts common LEDs. Trust matters.

If I had a bulb that'd last for 1000 years and cost $1, I'd keep my mouth shut, and market them at about the same price as LEDs, and claim they'd last 20% longer, and back that up with a warranty. I might even artificially limit their lifetime to, say, 2x the lifetime of the LED (or try to cost reduce to get there). Then I'd invest a good percentage of revenue into marketing, hopefully managing to outspend competitors on sales and marketing unit by unit.

Whenever a competitor would improve length, I'd then release a "new model" maintaining the lead. But not immediately, and never with too much of a lead.

I learned the hard way not to undercut / overcommit vs. your competitors with my first business: An ISP. We realised we could sell subscriptions at 25% what the incumbents could. What we did not realise at the time was why the incumbents were so expensive: It was not that they could not promise what we did for the price we did. It was that they did not see the mass market promise, and their margins were great as they were. So what happened was that 4 days after we launched, together with one or two other small ISPs that had also found the same price spot, the biggest ISPs in our market followed suit instantly, and started a price war that totally ravaged the market over the following 5 years. Nobody made any money until the major players had exhausted their war chests and carved up the market between themselves.

What I learned was that 1) you should not assume your lead is as great as you think; assume the competitors either have stuff hidden away or at least have resources to quickly cut your lead; 2) you should not assume your competitors can't cut price, so don't underprice them enough to force them into a price war with you; unless you have deep pockets, you will lose, 3) you should not assume your potential customers will find you just because of price (and if they do,you'll lose them the moment someone underbid you). In general, assume that you upset the status quo at your peril - everyone will gun for you if you reveal your hand and demonstrate a too large of an apparent advantage. In well established market with large incumbents, unless you have massive capital behind you, it's a bit like walking up to a sleeping tiger and giving it a swift kick: not very conducive to your long term survival.

It's better to come in under the radar. If you still want to change the world and destroy your market, do it once you've built up distribution and a war chest.

I found your comment an interesting read, and I want to think about it some more. However this one part struck me as why I wanted to comment earlier, " assume that you upset the status quo at your peril ", this is the exact opposite strategy to "disrupt everything" that comes from silicon valley, so I think there is a bit of a sw/manufacturing ideology gap. I suspect neither side is 100% wrong, more disruption chances could success in manufacturing, some software should take a more conservative stance off the bat, but I think where your speaking from, we don't hear enough about.

The Fairbanks Scales in the 19th century were famously reliable and durable. Sadly, for the Fairbanks, to the point that once one was bought, there was little need to replace it. The Fairbanks brothers (who lived in Vermont, not Alaska!) became fabulously wealthy. But their company is largely forgotten, due to it saturation the scale market within a generation or so....

I bought new bathroom scales about 5 years ago or so. The company I bought it from pretty much only sells scales, and they hilariously keeps trying to send me offers for discounted scales every few months. I've not unsubscribed almost entirely out of amusement: If I was in the market for another scale already, it'd mean theirs were of poor enough quality I wouldn't buy a second one from them.

It's unlikely that 1000 year bulbs would actually be useful for 1000 years. Just as incandescents are being replaced for sake of energy efficiency. And I find it hard to imagine any economic system that could efficiently build 1000 year bulbs. It requires spinning up factories for a short period of high volume, mass production, followed by a long, low volume trailoff. That's a costly and wasteful process under any economic system. Our limited time and money is better spent solving more immediate problems, methinks.

I think the original argument would have stood if the numbers were 20 to 50 years, and anything less than $1/yr or so.

Icandescents are being replaced because they were made illegal. If people really wanted the CFLs that would not have been necessary.

If it were possible, say, to create a light bulb that lasted 1,000 years for a dollar. Who could afford to make such a thing?

I sense that you think manufacturers don't have an incentive to produce quality products like bulbs that can be used for a long time. I think you're making a mistake which I'll try to illustrate with an example.

Suppose light bulbs cost 1 USD each and last 1 year. I invent a new light bulb that is the same except it lasts 2 years. Would you say selling it would be a bad deal for me? Of course if I were to sell it for 1 USD then it certainly would be. But the value of my bulb to consumers is at least 2 USD, because it's equivalent to buying and using two competitors' bulbs! Manufacturing costs aren't relevant - if my bulb is cheaper to produce then it's even better for me.

You shouldn't look at the cost of production because it isn't the only factor. Take a look at e-books, for example. Sometimes they're even more expensive than paperbacks!

> I sense that you think manufacturers don't have an incentive to produce quality products like bulbs that can be used for a long time.

They do, to an extent, if they're not a dominant market player. But the article demonstrates quite clearly that once you reach a certain size, the effect on overall market demand starts to become a concern that relates directly to how much product you'll expect to be able to shift.

> But the value of my bulb to consumers is at least 2 USD, because it's equivalent to buying and using two competitors' bulbs!

It's not, because people (rationally) discount future events: We won't fully believe the claim; we'll assume other things may make it break in the meantime; we may have more immediate uses for the extra dollar. A dollar saved a year from now is not generally worth a dollar today.

It will be worth more than the competing products (all else being equal), but it won't be worth twice just because it lasts twice as long.

It will be worth more than the competing products (all else being equal), but it won't be worth twice just because it lasts twice as long.

Absolutely agree. You rightfully point out broad simplifications that I made, though I think we agree that there's still an incentive for companies to build better products. Even if the price of the new bulb lower than 2 USD, but above 1 USD, the produces has an incentive to sell it - lower sales costs (you need to do it once), lower warehousing costs, etc. On the other hand the new technology can be more expensive than the old and offset these benefits.

Someone like Elon Musk would do it.

Some people are happy to make enough money to generate a comfortable (but not outrageous) salary for all involved for a limited period, if it means making the world a better place. Not everyone wants to screw over their fellow humans to extract as much rent as physically possible for absolutely as long as they can.

Sure, you won't raise much capital from Wall St. to do it, so you'll need a rich backer with philanthropic tendancies. They're a rare breed, but they do exist.

Most competently-constructed buildings and bridges today could last for a thousand years. They still get built, and they take a lot more upfront capital investment than any light bulb.

My guess: Buildings tend to be over-engineered because any failure will kill people. There's just no reason to buy a light bulb that will last longer than your own occupancy of a given place.

> Most competently-constructed buildings and bridges today could last for a thousand years.

That's not even remotely true. Without continual maintenance most of them would collapse in 100 years, maybe 200 on the outside. But that's not really the point. The real estate market is quite a different beast driven primarily by scarcity of space and location desirability.

I think there's a missing "'nt" in the previous comment.

A cartel is a way to constrain trade in an anti-competitive way. It is the opposite of capitalism. This is why antitrust laws and enforcement are so important.

In the zones with frequent natural disasters (earthquakes, hurricanes, ...) the replace rate will be higher.

Reply on news.ycombinator.com