Question: Suppose the demand for fentanyl is perfectly inelastic, and that the users of fentanyl steal from others to acquire the money to pay for it. In an effort to crack down on fentanyl use, the government imposes harsher penalties on suppliers of fentanyl, reducing its supply. How will this policy affect the amount of stealing by fentanyl users?
Solution: Students sometimes resist the idea that the law of demand applies to addictive drugs. I understand why: if something is addictive, it seems like people will consume it no matter the price. While that isn’t strictly true, let’s follow the assumption in the question that demand for fentanyl is perfectly inelastic.
What does that mean? Perfectly inelastic demand means the quantity consumed does not change when the price changes. Graphically, the demand curve is a vertical line. Users consume the same amount regardless of price.
Now consider what happens when the government cracks down on suppliers. The reduced supply does not change the equilibrium quantity consumed, since demand is perfectly inelastic. Instead, it raises the market price of fentanyl.
Here’s the important step: because users continue to buy the same quantity but at a higher price, their total spending on fentanyl rises. Since they finance their purchases through theft, this means the amount of stealing increases.
A simple numerical example makes this clear. Suppose a user buys one unit of fentanyl each week at $100, stealing $100 to do so. If supply restrictions raise the price to $150, the user still consumes one unit per week but now must steal $150. Stealing rises one-for-one with the higher price.
So, when demand is perfectly inelastic, reducing supply does not lower consumption. Instead, it raises prices—and in this case, leads to more stealing. If the assumption of inelastic demand were true, a more effective policy would target reducing demand rather than supply.
READER COMMENTS
Peter
Aug 18 2025 at 4:28pm
I just don’t think your last sentence, “a more effective policy would target reducing demand rather than supply.”, follows given the your constraint of a perfectly inelastic demand curve hence can’t meaningful affect demand though I’ll concede perfect to mean within normative ranges excluding at least the right side infinite money/harm tail.
I think you actually continue to build a case in your example in favor of something you were against in the discussion on your first post “With respect, I don’t think the question implies that we give fentanyl away. I certainly don’t favor such a policy!”. I think purely limiting it to your scenario, it does implies that completely as it’s an example of a market failure and regulatory capture. To use your numerical example, if one unit of fentanyl is $100 and the cost to society for that unit is $200 ($100 private party damages, $100 government tax expenditures for public safety to deal with it) but nothing we do within normative ranges is going to change the demand, if we can reduce the cost to free we end up with no additional users, a vast reduction in harm and government costs, and more liberty. Because all decreasing supply is going to do, to your point, is raise costs, both financially and in actual harm, for no gain and you cant’ affect demand. Basically I think you are proving a case for harm reduction but can’t square that in economic terms though I’d just coach it in trade offs, not SD curves.
I get there is no such thing as a free lunch but using open source casual material and real world lived experience, right now it’s about ten cents a dose to manufacture and that could come down vastly pulling it out of the black market. Given a daily usage rate of ten doses a day that’s about $365 dollars a year vs. $250 A NIGHT to jail them and that’s not even counting all private party damages from the crime incurred to get the money nor the secondary costs to arrest and prosecute them. So sure lunch isn’t free but it’s actually better, we are actually making money via vast savings while also being better people.
Jon Murphy
Aug 19 2025 at 7:41am
A perfectly inelastic demand curve means that quantity demanded does not adjust with price. The curve is vertical, so the same quantity is demanded at any price.
Reducing the demand means the entire curve shifts to the left. That will reduce quantity demanded, even though the curve is inelastic.
MarkW
Aug 19 2025 at 6:55pm
Fair enough. Maybe it would be clearer to say something like ‘reduce the quantity supplied at the old price as the supply curve is forced to shift to the left’. The inelastic demand means that the total quantity bought and sold would remain the same but at a new, higher price — meaning that addicts would have to steal that much more to afford their fixes.
Jon Murphy
Aug 19 2025 at 7:06pm
Yes. That is the correct answer
Peter
Aug 19 2025 at 7:09pm
Yep, that was my understanding to from the narrative Bryan gave, i.e. we were treating it like water, there is no way to reduce individual demand as you need a certain amount daily to live hence no substitution affects, no way to change demand, etc.
MarkW
Aug 19 2025 at 8:20am
reducing supply does not lower consumption
It sort of has to, doesn’t it? If supply is actually reduced by half, consumption will also necessarily fall by half (unless before, half of the supply was somehow discarded because it was past its expiration date).
More likely, it seems, that cracking down on suppliers would result in lots of seizures of drugs, but because the difficulty and risk of supplying the drugs would go up, so would the price and rewards for successfully evading the authorities and delivering the product. So addicts would need more cash and steal more to buy the same amounts as before the crackdown.
Jon Murphy
Aug 19 2025 at 8:46am
Recall that reduced supply is not the same as reducing quantity supplied. Reducing supply means a fall in the supply curve (a shift to the left). That, in turn, means that at any given price, producers offer fewer units (or, what comes to the same thing, producers must now be paid more to offer the same amount as before).
With a perfectly inelastic demand curve, as supply falls, quantity demanded stays the same, and so long as the two curves continue to intersect, the same amount will be exchanged in the marketplace.
Now, if supply completely disappears, then yes the market will disappear as well. But those edge cases are extraordinarily unlikely, especially with an inelastic demand curve. You’re more likely to create monopolies (ie drug cartels) and deeper black markets.
David Seltzer
Aug 19 2025 at 1:16pm
Jon wrote; “Recall that reduced supply is not the same as reducing quantity supplied.”
An example: In 1956, a collector purchased Picasso’s a 15-piece suite of “Femmes d’Alger” series (1954–55) for a reported $212,500. In years to follow, The collector recouped his initial investment buy selling off ten. Version ‘O’ was among the few he kept. Version ‘O’ of “Femmes” remained with a collector until it was auctioned by Christies for $179.4 million, May 2015. As supply for this piece is fixed, the auction tells us the price of Version ‘O’ was bid up.
Warren Platts
Aug 19 2025 at 12:12pm
What if the interdiction is so effective that it puts a hard lid on the quantity of fentanyl supplied rather than merely reducing the supply? That is, the interdiction effectively renders the supply perfectly inelastic at a point that is much less than the quantity demanded? No doubt the street price would increase, but many users will not be able get their fix at any price. What will they do?
I see two options: (1) so many users go straight such that the demand for fentanyl would become elastic and thus the quantity demanded would match the quantity supplied; or (2) users switch to domestic suppliers, as happened with both cannabis and methamphetamine. Assuming would-be suppliers can’t get the right precursors, then they might turn to things like growing their own opium poppies. (In fact, it is technically legal in the USA to own poppy seeds and even grow them as long as you don’t make opium out of them.)
john hare
Aug 19 2025 at 5:57pm
Your last sentence reminds me of a sign posted by a hunter.
BAITING DEER IS ILLEGAL
This corn is for the chipmunks, squirrels, and rabbits.
Any deer caught eating this corn will be shot.
Jon Murphy
Aug 19 2025 at 6:56pm
A few notes:
1) your first option wouldn’t occur. If users “go straight” then the demand curve would fall, but it wouldn’t go more elastic. Recall that the number of buyers in the market affects the position of the market demand curve, but not its shape. (This is easily demonstrable even if one doesn’t know the logic behind the models. Just look at how market demand curves are calculated and how elasticity is calculated and see which would be affected by a decline in buyers).
2) your second option violates the condition of your hypothetical. If there are substitutes to domestic suppliers (whatever that’s supposed to mean. The question doesn’t specify a crackdown on just foreign fentanyl), then the supply curve is necessarily relatively elastic, not inelastic. (Again this is easily demonstrable by looking at the determinants of elasticity of supply).
As a general rule and what I tell my students: it is better to be right then clever. Look at what the question is asking and focus just on what the question is asking. If you find yourself importing all sorts of assumptions, you’re changing the question (likely to justify a previously determined answer, either consciously or unconsciously) and thus will get it wrong.
Mactoul
Aug 20 2025 at 1:15am
Shouldn’t it be reducing the quantity supplied?
Does govt action intend at supply or quantity supplied?
Or the idea is that whatever the govt intends, it can only reduced supply but not the quantity supplied?
Jon Murphy
Aug 20 2025 at 7:28am
No. Supply falls since the cost of producing rises due to the government involvement. But since demand is perfectly inelastic, the quantity supplied does not fall. It stays the same. Only the price rises, which in turn triggers the higher crime rates.
Mactoul
Aug 20 2025 at 9:54pm
Are you saying that the government interdiction of drugs has no effect other than raising the price?
By what mechanism does govt does that except for the normal way of lower supply–> higher price?
By inelastic demand, I understand that the customer buys the same quantity whatever the price might be. But how does this customer preference cause the amount of drug to appear in the market?
To me it seems like reasoning from the equations and ignoring causality.
Govt has casual influence on the amount of drug reaching the market but that precisely you appear to deny.
Customer preference ie inelastic demand, to my mind, cannot have this casual influence on the amount of drug reaching the market, but you appear to say it is the only thing influencing the amount of drug reaching the market.
Jon Murphy
Aug 21 2025 at 7:51am
If addiction implies a perfectly inelastic demand curve and the interdiction is against suppliers, yes.
If one wants a more effective policy to reduce drug use, this model suggests that it’d be more effective to go after demand, not supply (eg rather than busting up drug dealers, use programs like DARE).
Mactoul
Aug 22 2025 at 2:59am
If demand is perfectly elastic and final price not infinite, that I suppose government intervention did not manage to reduce the quantity of drug reaching the market.
Since, it is not realistic to assume that govt interdiction has no effect on the quantity of drug reaching the market–at least some drug must be intercepted.
Otherwise, one may assume that the suppliers increase the pre-intercept supply such that the same amount is reaching the market.
Maybe the assumption of perfect elasticity, while good for theory, is not realistic enough to draw the policy — it’d be more effective to go after demand, not supply.
Jon Murphy
Aug 22 2025 at 5:51am
If demand were perfectly elastic, then going after suppliers would reduce quantity in the market (but not the price). But that’s a different question than the one Bryan asked.
Mactoul
Aug 23 2025 at 12:30am
Sorry I meant to write “if demand is perfectly inelastic” above, not elastic.
Jon Murphy
Aug 23 2025 at 4:43am
In that case, we need to rework your comment a little. You keep talking about assumptions that aren’t there:
Those are not assumptions. Those are results. We see that, assuming demand is perfectly inelastic, that interdiction merely raises the price but doesn’t affect the quantity supplied.
Perhaps, but you need to give a better reason than the one you have, which is just that you don’t like the results of the analysis.
Comments are closed.