Let's say you see a potential pandemic coming, and you produce a product that could be critical. Maybe you make respirator masks, maybe you make ventilators, maybe you make PCR test reagents. You can see that if you and your competitors don't ramp up production and the pandemic happens, there will be a shortage. What do you do?

One option is to do nothing: keep producing at your regular rate. If the pandemic fears were overblown then you're fine. If the pandemic happens you quickly sell out, and start scrambling to ramp up production.

Another option is to ramp up production now, speculatively. Start paying workers extra to work longer shifts and run your assembly lines around the clock. Train extra workers. Find what you're bottlenecked on and figure out how to get that ramped up too. If the pandemic fears were overblown you lose a lot of money, but if the pandemic happens people need what you have so much that you can charge high prices. How much to ramp up production in advance depends on how likely you think the pandemic is, and how much you'd be able to increase prices if it does happen.

Except we have laws and customs against price gouging: if the pandemic does happen, you are going to have a lot of trouble raising your prices. The laws generally do allow passing along increased costs, but the problem here is that your costs were speculative. Let's work an example.

Imagine your ventilators normally cost $35k to make, and the market price is $40k each. If you push really hard to ramp up production you can make a lot more, but your cost goes up to $90k/each. In New Jersey, the state I looked at last time, you're allowed to pass along costs but can't increase your profit on "merchandise which is consumed or used as a direct result of an emergency or which is consumed or used to preserve, protect, or sustain the life, health, safety or comfort of persons or their property" by more than 10% (56:8-107, 108, 109). Your normal profit is 14%, so you would be allowed to sell them for $104k. Your possibilities are:

  • No pandemic: you spent $90k each, but the market price is $40k. You lose $50k each.

  • Yes pandemic: you spent $90k each, and the market price is way above that, but you can legally only charge $104k. You make $14k each.

If you think the pandemic is >78% likely to happen then you'll expect to make money by ramping up production, otherwise you'll lose money. So even if a pandemic looks, say, 75% likely, you don't ramp up.

Similarly, imagine you make respirator masks and you know that every so often there's an emergency where demand spikes. Could be a pandemic, but could also be widespread fires, or many other things. Since melt-blown fabric is a major bottleneck, you could decide to keep a large stockpile of it so you can easily ramp up production in an emergency. The same unfavorable math applies here: you're heavily limited in how much you can increase prices in an emergency, so keeping a large stockpile to be prepared for even a reasonably likely event is a money-loosing proposition.

In the current crisis people are likely to die because we don't have enough ventilators, and the marginal person probably needs one for about a week, and the peak lasts maybe two months, so the marginal ventilator saves about eight lives. At the US statistical value of life of ~$9M, that's $72M per ventilator. We're heading into a disaster where we don't have enough machines that we would value at ~$72M/each and normally cost ~$35k/each to make. This is really bad.

It's too late to fix this for the current situation, but I see three main ways out of this for the future:

  • Allow price gouging: don't restrict what prices people can sell things at.

  • Allow speculative production: require companies to disclose and document production plans, require them to share their probability estimates of how likely they think things are to be needed, keep them honest by allowing third parties to bet against them at their published probabilities.

  • Have a government that will put in emergency ventilator orders at an early stage of the crisis even when they may not be needed, stockpile masks for potential pandemics, and generally stay on top of things.

These three approaches require increasing levels of government competence and foresight, and given recent performance I'm pretty skeptical. But the current approach where the government doesn't handle the problem and also does not allow industry to make a profit handling the problem is a disaster.

Comment via: facebook

New Comment
16 comments, sorted by Click to highlight new comments since: Today at 4:12 AM

Go to the bank and tell them "I need a contract that will pay out money if there is no pandemic.". (The bank is now rubbing their hands, because this offsets their risk.) Your costs are no longer speculative, and you can safely pass on the cost of the contract to the consumer.

That's a much better version than my "require companies to disclose and document production plans, require them to share their probability estimates of how likely they think things are to be needed, keep them honest by allowing third parties to bet against them at their published probabilities".

This is a lot like a pandemic bond, though apparently they existing ones still haven't paid out? https://apnews.com/9b6191d5d82145c237b470525e2e9515 https://www.npr.org/2020/03/24/821000049/pandemic-bonds

The fundamental problem that businesses solve is that marginal cost is different from average cost (usually lower, in this case higher). One standard way to address it is price discrimination - find a way to get more willing customers to pay more, by bundling, by slight variation in quality for larger variation in price, or by other things that matter more to customers than to costs.

The obvious business advice for such a manufacturer would be to introduce a new line of premium ventilators, which are mostly the same, but have copper touch surfaces or something. Do this long before any hint of specific emergency - this is normal business practice. Price them at $100K normally, and you probably won't sell many. Start ramping these up when you predict a surge, and these will be the ones you sell when you're out of stock of the cheaper ones.

Oh, don't forget discount policies - you should normally have a list price of $90K, and discount that to various levels based on negotiated delivery and volume commitment. You don't have to offer discounts for last-minute purchases that weren't contracted 6 months ago.

(all of these fall under the heading of "allow price gouging". That's the right answer for production-side incentives. It's the WRONG answer for speculators who amplify the shortage by taking away from normal supply, rather than providing new supply. And it's deucedly hard to tell the difference from outside, so governments tend to just outlaw the whole idea.

Is there a way to do this if you're not already a manufacturer? With, say, N95 masks--could we buy up a bunch of regular ones when there's no shortage, do something that costs a little money and adds a tiny bit of value (maybe print cute pictures on them for doctors who wear them around little kids) and then mark them up 900%? I mean, I don't have money to buy them with, but if stockpiling really makes economic sense then I'd think we could get investors to front it.

How do price gouging laws apply to discounted products? If we almost always offer an 80% discount, can we stop offering that during an emergency?

ETA: to be clear, masks are just an example. They would not be a good choice in the future because after this pandemic governments will actually stockpile them. We'd need to figure out what there's likely to be a major shortage of in the NEXT global emergency.

you're not already a manufacturer

Sure, it's common for a reseller to bundle support, installation services, or warranty with items, in order to justify very large markups. The discount methods work too. A packaging or product change is closer to the manufacturer case - you have to deal with time/cost/capital tradeoffs to change your production rate or inventory size.

As a pure speculator, you're starting off in a worse spot, because you're dependent on the manufacturer(s) for supply, and you'll start with slimmer profit margins, which makes your risk much higher if the shortage doesn't occur as severely as you expected, or if manufacturers can ramp up faster than you predicted.

You're also in a morally worse spot - you run the risk that you are creating or amplifying the shortage, not just predicting and smoothing the consumption of rapidly-value-changing products, and not actually increasing supply with the increasing price. If you aren't solving the problem given in the post (you could make more, but not at your normal prices), you're not as clearly on the side of good.

Keep in mind that simple models are pretty misleading. There are almost NO manufacturers who aren't also middlemen for parts of their equipment. They don't dig up and smelt their own ore, they buy everything (often significant manufactured subcomponents) from other manufacturers. There are few distributors who aren't adding real value with some amount of warranty/returnability, and delivery or quantity options that the manufacturer doesn't support. These things really blur the line between producer and middleman.

You're also in a morally worse spot - you run the risk that you are creating or amplifying the shortage, not just predicting and smoothing the consumption of rapidly-value-changing products. If you aren't solving the problem given in the post (you could make more, but not at your normal prices), you're not as clearly on the side of good.

I don't follow. The idea would be that you aren't contributing to the normal supply, you just stockpile in case of future emergency. This increases production during normal times slightly, and then when there is a pandemic you would stop purchasing and start selling. This reduces the need for manufacturers to produce more at higher marginal cost. How could it create or amplify a shortage?

It's true you'd have no profit prior to the emergency (and large costs) but in theory that shouldn't matter as long as your investors diversify.

Stockpiling over time, long in advance of a crisis is great (but might be tough to find those investors).
Supply has time to adjust and you are not hurting anyone.

Stockpiling in a burst just before a rush, when it's too late for suppliers to adjust production, is much less clear. You're accelerating or in some cases creating the shortage, and you're not changing the short-term supply at all.

Ah, okay. I guess that answers my question: you don't think it's possible to mitigate shortages by stockpiling without losing money in expectation. I wish it were--surge production is a much more tenuous solution that depends on being able to foresee a disaster shortly before it occurs, and even then it might not be possible to produce enough to help significantly.

(all of these fall under the heading of "allow price gouging". That's the right answer for production-side incentives. It's the WRONG answer for speculators ...

I'm not sure that's true. Speculators are an important part of price discovery. The problem is that lots of people hate what they 'discover' and thus we generally fail to heed those discoveries.

The general problem seems to be (rapid or extremely rapid) scalability and that seems like just a very hard problem to solve.

Within normal constraints and timeframes, this is absolutely true - speculators are performing an important role in predicting and propagating demand. For short-term events which are ALREADY visible to suppliers (and somewhat less so to consumers) speculation has less information value, and more disruptive impact by their normal-value of shifting demand forward. In these situations, they shift a temporary demand into a much higher peak than it otherwise would be.

In other words, speculators are valuable usually, because usually they're flattening the curve. They're harmful in some exceptional circumstances because they cause the spike to be artificially higher than it otherwise would be.

Note that it's NEVER this black and white. It's certainly true that there are both benefits and harms to different constituencies in all cases, and where you net out will depend on what you're focusing on.

That's a fantastic point and one that Alex Tabbarok made in this post recently – 'ticket scalping' basically.

There is nothing stopping customers from buying such supplies during good times. The buyer in option 3 doesn't need to be a goverment. But if you have a supplier that does do emergency production and a supplier that doesn't then chances are the on-demand supplier prices are cheaper. A homo economicus could factor delivery dependence to what they are willing to pay. But a real human migth overweight an extra delux food item over nebulous future happenings.

US dismantled its pandemic unit because "it was costing a lot of money" and "we haven't used it for years".

Rewarding rushed production also rewards inefficiency. If you stockpile the supplies you can use the low $ per item. Delaying production keeps the resources open to be used and until we have more threats than we can predict that is not a very useful property.

So is there a way to run a charitable organization that accomplishes this?

For example, if an org started ordering respirators to be built to expand capacity starting in January, would that be a good fit for Open Phil funding?

Would LW be able to convince people to move money to it at that point?

There is the style where those who participate in the preparing benenfit from the preparing. A kind of preppers health insurance.

And then there is the version where you would let randoms that have not participated in such bunkers should they need to be used.

You could have middle forms where you don't require full compensation for the service or let it be contingent on the world not ending etc. You could just make the d equpiment, hand it out, write some sum to be collected in 10 years. Then in normal conditions if you let the pieces of paper lose their meaning it was totally an act of charity, if you expect 100% of them back it's full price gouging. but you could also count on a partial public payout ie goverment pays whatever it feels like at that time like 15% of the nominal sum as actual money because it would look PR bad to not pay anything kind of like how war debts often get partially forgiven.

It would count on people accepting risk of not getting a properly legally binding assurance just the problems starts to get fixed.

As an aside, I'd push back gently against using a dollar figure for the statistical value of a human life to set the value on any single life-saving (or life-sustaining) product. I may be misunderstanding the concept, but it seems like assigning the entire value of a saved life to one product used in one instance would be a mistake.

My thought process: If every additional tool used to save any life (or the same life multiple times) is valued at the full $9 million per each life saved, then the actual "value" of a life would be many, many, many multiples of $9 million.

All that said, it is interesting to read this post in light of efforts to ramp up production of ventilators and masks.

That's a good point, but I don't think it changes the analysis much. All of the "additional" tools are already available (in some sense and to some degree) so those costs are effectively 'sunk'. The tools or supplies that are necessary to save a life, but are unavailable, effectively 'swallow' the entire value of the lives they could save. This is definitely 'marginal' thinking ('thinking on the margin').

But your point is still true to an extent even in the case of ventilators – it's not sufficient for someone to have produce a ventilator – somewhere. It also has to be delivered, installed, and then run – by certain specific people with the relevant training and credentials – and, in the medium and long terms, inspected and maintained. Even on the margin, all of those contribute to the value of any lives saved.