Why Don't The Airlines Just Raise Fares?

[mweiss]

I believe I read on the ATA website about productivity, 30% since the mid eighties comes to mind, and no one can dispute the huge increases in fuel efficiency.

Sorry, I thought you meant operational efficiency. While it's true that, industry-wide, operational efficiency has increased over the past two decades, you can give nearly all of the credit to the growth of Southwest. Look at the CASM over time for all airlines, including Southwest. It has increased faster than inflation.

By the way I'm still waiting for you to prove to me that the airlines cant raise fares.

I hope to get there. But you're fighting me every step of the way.

Back in the eighties B-scale and Lorenzos famous trip to BK court were not attempts to increase revenue but cut costs.

Lorenzo was way too simplistic. The only two ways he could figure out to cut costs were to cut wages and get rid of services. It's not an effective long-term strategy.

I've been in this industry for over twenty five years, nearly all that time there has been an obsession with cutting labor costs and the airlines have been very successful at it.

Because it's easier than improving operational efficiency.

Sure but did you se the relationship between these lean years and the extreme booms that follow once the concessions are won from the workers?

Yes, but the concessions aren't the reason. Believe it or not, the economy drives airline profitability, not the other way around.

And you are assuming that there is only one dimension to what is going on in this industry.

Not at all. I'm only trying to get you to sit still long enough to start with a single dimension and grow the discussion from there. The way to eat an elephant is one bite at a time.

 

[mweiss]

The answer to (who are those that are not trying to break labors back and seem to have a YM that is functioning correctly) is obvious.

Humor me and tell me who it is. I know the RASM of every major airline in the US, so I'm dying to know who you think is doing the best job at yield management.

Does [Southwest] have lower labor costs? Prove it.

Look at their labor CASM. It's lower than every legacy carrier.

Ive seen figures to the contrary.

And those figures are...?

Ok, but [Southwest's employees] are not moving them to Tokyo or London, markets where passengers pay a lot more money.

Um...that's revenue, not cost.

The total cost per passenger mile drops as the length of the trip grows, or at least it should. Are you switching from comparing rpms to numbers of passengers to suit your arguement?

No. I misspoke, and should have said that the number of RPMs served per employee is higher at WN...which makes the legacies look worse.

We are still at the "lots of start ups stage".

No, now we're at the "occasional newcomers" stage. You can't compare three new airlines to the free-for-all that we had in the early 80s. 🙄

Somehow, I believe they will find [fighting cash]. What is GE going to do with all those planes otherwise?

I hear there are other countries outside the US that have air travel.

Perhaps they have been taking defensive actions against other carriers while at the same time taking agressive offensive action against their own employees.

They're not selling air travel to their own employees. We're talking about setting ticket prices.

But your stance is becoming clear. Your point is that the legacies are poisoning themselves just to stick it to labor.

 

[mweiss]

Fine. let SWA take over all the short haul domestic routes.

That was pretty much US's strategy. But now WN is also moving into domestic long haul. And, anyway, domestic short haul remains necessary to provide feed for international in a profitable fashion. Those pincers are closing.

Maybe you need to go back a few more years to get the clarity that airline workers who have been here have. Then maybe you would understand our skepticism.

I understand your skepticism. I'm just telling you why it's different this time. Recall in the story that eventually the wolf did show up.

We have seen these cycles before, and the same rhetoric gets spuced up and slightly altered each time.

This is not the same thing. In previous economic cycles, the downturn reduced every airline's capacity. This time, the LCCs grew during the recession. It's a significant difference that shouldn't be taken lightly.

The LCCS will die off or be bought off

This time, the LCCs are in a position to buy the legacies, rather than the other way around. Not that they will; it's a much better strategy to wait for the legacies to die. This is not the same landscape as 1986 or 1991.

, the industry will continue to consolidate and at least one of the legacies will dissapear. Then like a pheonix from the ashes those that remain will have a miraculous recovery and boast of record profits. In the meantime the employees will be stuck with concessionary agreements that last into the next decade.

A conspiracy? Sure why not? ...there is plenty of motive and everyone other than the workers stand to gain from lower labor costs...savings in labor can be diverted to several other benificiaries, those who do business with the airlines can charge more (we saw our concessions get eaten up by fuel)

Um...yeah...the price of oil shot up because a handful of airlines in the US managed to get concessions from labor. 🙄

 

[Bob Owens]

mweiss,Jan 9 2005, 06:54 AM]
Yes, but the concessions aren't the reason. Believe it or not, the economy drives airline profitability, not the other way around.

So we agree that our concessions for the last twenty years were pointless?

The way to eat an elephant is one bite at a time.

Yea but first you have to kill it.

 

[Bob Owens]

mweiss,Jan 9 2005, 07:09 AM]
Humor me and tell me who it is.

Perhaps if I were posting under an alias I would, you're a smart guy, I'm sure you know who I mean.

I know the RASM of every major airline in the US, so I'm dying to know who you think is doing the best job at yield management.

Who is makling the most money?

Look at their labor CASM. It's lower than every legacy carrier.

Yea but you have to look at their total CASM. SWA contracts out a lot of their work, so those costs would not be considered labor cost like the work done in house. The costs are still there, just in a different catagory. UAL for example contracted out a lot of maintenance so their labor costs for maintence went down while their total costs for maintenance went up. SWA singular fleet also helps lower their costs. However such a strategy is not practical for long haul markets.


Um...that's revenue, not cost.

True, but if we are serving different markets that generate higher revenue then competing cost wise is not as critical, if you are earning higher revenues you can absorb higher costs.

No. I misspoke, and should have said that the number of RPMs served per employee is higher at WN...which makes the legacies look worse.

Not really. SWA contracts out a lot of work, so their figures per employee would be higher but it does not reflect the total amount of costs that were absorbed by people working to provide the service because many of those people would not be employees of SWA.

Peoples Express is an example of an airline that had the lowest maintenance labor costs per RPM, but that was because they contracted it all out. The large group of inexperienced mechanics over there did a tremendous amount of damage in a very short time including but not limited to running an airplane off the runway and crashing it during a high speed taxi-because they had the stab trim improperly positioned(when the nose unexpectedly lifted off the ground they panicked), bending an airplane-yes bending it-because of improper jacking and having an airplane roll into a ditch because they did not chock it right.

So while Peoples could post some impressive numbers not only were they misleading but they also hid a lot of other costs.

No, now we're at the "occasional newcomers" stage. You can't compare three new airlines to the free-for-all that we had in the early 80s.

Sure we can, the newcomers are better funded and there are fewer legacies.

I hear there are other countries outside the US that have air travel.

Yea and most of them dont have much money. That old demand issue pops up again, people who live in a straw and mud hut and can hardly afford food are not likely to be jetting off on vacation.There are plenty of stage II aircraft that American companies are already leasing to those countries and there are a lot out in the desert already. In fact a piece of one of them may be coming to an office near you! From what I hear they are running out of room to put them.

They're not selling air travel to their own employees. We're talking about setting ticket prices.

Well the issues are related. They claim they must cut our wages because they can only sell low priced tickets.

But your stance is becoming clear. Your point is that the legacies are poisoning themselves just to stick it to labor.

No they are doing it because they are greedy, nothing personal, just business.

 

[Bob Owens]

mweiss,Jan 9 2005, 07:51 AM]
That was pretty much US's strategy. But now WN is also moving into domestic long haul. And, anyway, domestic short haul remains necessary to provide feed for international in a profitable fashion. Those pincers are closing.

Thats what we have Eagle for, besides is it really necessary, or is it advantageous to have feeder for International?

I understand your skepticism. I'm just telling you why it's different this time. Recall in the story that eventually the wolf did show up.

Well if the wolf offers better pay and benifits then by all means let him in.

This is not the same thing.

It never is.

In previous economic cycles, the downturn reduced every airline's capacity. This time, the LCCs grew during the recession. It's a significant difference that shouldn't be taken lightly.


Not true, I worked at two LCCs that grew while the majors shrunk in the early 80s.

This time, the LCCs are in a position to buy the legacies, rather than the other way around.

Great, the sooner the better. The more info you give the more convinved I am that concessions are not the way to go.

Not that they will; it's a much better strategy to wait for the legacies to die. This is not the same landscape as 1986 or 1991.

I was referring to 1981-3, but are you saying that the majors shrunk in 1986? AA started a huge hiring spree in 1986.


Um...yeah...the price of oil shot up because a handful of airlines in the US managed to get concessions from labor.

Not exactly, a businessman exploits the opportunities presented to him whether they just happen or are created but the fact fuel consumed all our concessions only proved that our concessions did nothing to better insure our financial future.

The best thing that we could have done is to collectively withdraw our labor and shut the air transportation system down. This should have been presented to the public the first time an airline claimed that they were going to a judge to get a labor contract abrogated or modified.

Once the overwhelming majority of the Air TRansportation system was shut down then other parties that are effected by the loss of service would pressure the government to do something in order to get it running again. We could then make an arguement from a stronger position that we are not going to see our wages slashed in order to save dying companies, if they cant make money then shut them down, the labor will be needed somewhere to provide the service or if the public is not willing to pay the price for the service then eliminate it.Either way, we move on.

The fact is the airlines are an essentail service and the airlines move more than just the discretionary vacation traveller or the dissapearing business traveller. Its importance is cited when workers threaten to withdraw their labor and the government steps in, well how can they on the one hand say that we are private workers but we have a public duty and they will help to lower our pay but block any attempts to raise it? How can they say that the public has a right to cheap air travel but those who provide the labor to make that travel safe and relaible dont have the right to a fair wage?

 

[mweiss]

So we agree that our concessions for the last twenty years were pointless?

[post="237132"][/post]​

Some were, some were less so. Personally, I think the unions leadership didn't negotiate contracts properly to recognize the cyclical nature of the industry.

I'm dying to know who you think is doing the best job at yield management.

Who is makling the most money?

In revenue, that'd be US.

Yea but you have to look at their total CASM.

Fine. Their total CASM is close to the legacies' labor CASM. Somehow they're still managing to pay for gate leases, landing fees, fuel... And, incidentally, they don't contract out much beyond the overhauls, which I understand they're considering bringing back in house. Furthermore, every airline is contracting out; WN actually keeps more of their total work in-house than most legacies.

SWA singular fleet also helps lower their costs. However such a strategy is not practical for long haul markets.

So? Where did I suggest that it was? You're moving into costs, which have nothing, nada, zero to do with the demand for air travel.

if we are serving different markets that generate higher revenue then competing cost wise is not as critical, if you are earning higher revenues you can absorb higher costs.

This thread is not about profits. It's not about costs. It's about revenues. How about we get back to revenues. If you want to debate costs, by all means start a new thread and I'll be there.

 

[mweiss]

You can't compare three new airlines to the free-for-all that we had in the early 80s.

Sure we can, the newcomers are better funded and there are fewer legacies.

Which is precisely what makes it an apples-to-oranges comparison.

That old demand issue pops up again, people who live in a straw and mud hut and can hardly afford food are not likely to be jetting off on vacation.

Wow. I had no idea that the US was the only country with houses made out of materials besides straw and mud. You sure showed me.

There are plenty of stage II aircraft that American companies are already leasing to those countries and there are a lot out in the desert already.

Well, yeah. Stage II aircraft are outdated. Most of the civilized world won't take them. What's your point? Do you think that if US Airways folded tomorrow that their Stage III aircraft would suddenly lose an "I"?

They claim they must cut our wages because they can only sell low priced tickets.

So what? This thread is not about your wages. It's about pricing and associated revenue. Stick to the topic, at least until we reach a conclusion.

Thats what we have Eagle for (international feed), besides is it really necessary, or is it advantageous to have feeder for International?

It is advantageous at the least, and necessary at the most. But that's for another thread.

Not true, I worked at two LCCs that grew while the majors shrunk in the early 80s.

I don't count the early 80s because there wasn't enough time for the fundamental effects of removing route barriers to take effect. That's why there's not much point in looking further back than about 1988, which was the end of the post-deregulation consolidation cycle.

So...getting back on topic here...do you agree with the theory of optimal pricing? That is, specifically, do you agree that if you know the shape of the demand curve that you can determine the optimal price to maximize revenue? Further, do you agree that if you know both the shape of the demand curve and the shape of your marginal cost curve that you can determine the optimal price to maximize marginal profit?

Once we're past this, then we can discuss whether it's possible to know those things. Eventually, I hope we can talk about the bigger stuff that you keep coming back to.

 

[Bob Owens]

mweiss,Jan 9 2005, 06:33 PM]
Well, yeah. Stage II aircraft are outdated. Most of the civilized world won't take them. What's your point? Do you think that if US Airways folded tomorrow that their Stage III aircraft would suddenly lose an "I"?

No but a lot of them would likely end up in the desert. How much would GE be collecting in leases from an aircraft sitting in the desert?

So what? This thread is not about your wages. It's about pricing and associated revenue. Stick to the topic, at least until we reach a conclusion.

Well who made you the moderator? Dont you mean your conclusion?

I don't count the early 80s because there wasn't enough time for the fundamental effects of removing route barriers to take effect.

Tell that to the thousands of employees that were effected.

That's why there's not much point in looking further back than about 1988, which was the end of the post-deregulation consolidation cycle.

Are you saying that consolidation is no longer the industry trend?

So...getting back on topic here...do you agree with the theory of optimal pricing?

Its not whether I agree with it or not, its whether or not that is the driving force of pricing. I feel that its not.

That is, specifically, do you agree that if you know the shape of the demand curve that you can determine the optimal price to maximize revenue?


When all considerations are factored in.

Further, do you agree that if you know both the shape of the demand curve and the shape of your marginal cost curve that you can determine the optimal price to maximize marginal profit?

Sure, like you said, if.

Once we're past this, then we can discuss whether it's possible to know those things. Eventually, I hope we can talk about the bigger stuff that you keep coming back to.

Ok. Lets go.

But I think we are going throiugh a lot of discussion to get to the point where we will disagree, that being what people are willing to pay for airfare and why the airlines are setting prices where they have.

 

[mweiss]

So, now that you've agreed that optimal pricing theory itself works, let's move on.

Do you agree that it's possible to determine, within a small margin of error, the shape of the demand curve for a given market?

Do you agree that the demand curve is clearly bimodal, albeit with different sizes of the two modes depending on the market?

Trust me, if we get this far, it'll go quickly from here to the meat of the discussion.

 

[Bob Owens]

So, now that you've agreed that optimal pricing theory itself works, let's move on.

Do you agree that it's possible to determine, within a small margin of error, the shape of the demand curve for a given market?

Do you agree that the demand curve is clearly bimodal, albeit with different sizes of the two modes depending on the market?

Trust me, if we get this far, it'll go quickly from here to the meat of the discussion.

[post="237337"][/post]​

Go ahead state your case.

 

[mweiss]

Knowing the shape of the demand curve for a given market is only enough when you have a monopoly. To date we've seen "optimal" pricing in the handful of monopoly markets out there. We've even seen a small number of "competitive" markets that are priced as if they were monopolies.

With a bimodal demand curve, you have a few choices in pricing. One is to choose to serve only the higher-end mode. The good news in going this route is that you don't have to worry about looking at the lower-end mode at all; you simply price everything at the high end, and customers self-select (i.e., leisure customers won't darken your doorstep). This approach is fine, but it limits your growth. If you're an airline, you probably have contracts with employees that mandate that their wages rise faster than their productivity. This means that the longer you grow more slowly than the wages rise, the greater your cost disadvantage.

You could try instead to go after the entire curve, which would allow for growth to outpace the difference between rising wages and rising productivity. But the problem with a bimodal demand curve is the need to accurately differentiate between customers in one mode and customers in the other. This was the impetus for saturday night stays, nonrefundable ticketing, change fees, and long advance purchases. What else can you use to tell whether you're selling to a business customer or a leisure customer?

This was the approach taken by the legacy carriers. Leisure travelers, naturally, have no incentive to look like business travelers, but business travelers have incentive to look like leisure travelers. Once the price differential grew beyond a certain threshold, the cost to businesses of masquerading as leisure travelers became lower than the cost of simply looking like business travelers. It became cheaper to pay for the extra hotel and car rental than to pay for a return on Friday. Business conferences, in order to encourage attendance, began offering events on Saturday nights. Businesses figured out that, even with a couple of changes to an itinerary, it still was cheaper to buy the leisure fares.

In other words, yield management got so good at addressing the first order need (i.e., maximizing revenue for each flight) that they created a second-order condition (i.e., a change in the behavior of business travelers).

There's a third choice that can be made. Airlines could price based on the lower mode only. They'd sell to the entire demand curve that way. The downside to such an approach is leaving money on the table. Optimum pricing theory would dictate that this is a bad idea. And, in fact, it is a bad idea under most circumstances.

There are plenty of hybrid versions that can precipitate from these models, and I'm not going to elaborate on them. They should be obvious.

Instead, I'm going to move to a discussion of where the optimal pricing theory breaks down. Or, more precisely, what needs to be done to expand it to encompass more scenarios. See, optimal pricing theory, as described over the past few days, assumes a monopoly. Monopolies are great fun in economics courses because they're easy to model. They're not, however, that common in business in the US, primarily because of laws passed about a century ago to encourage more universal economic improvement nationally.

So, we can't have true monopolies in this country (barring a few exceptions that can be saved for a different discussion). What, then, happens in a competitive marketplace? How does optimal pricing play out there?

Let's look at a simple example. Say Producer Alpha can make a widget for $1 each, and Producer Bravo can make the same widget at a cost of $2 each. If the demand curve in a monopoly market would dictate an optimal price of $4 (I'm making an assumption here that's inherently false, since Alpha's optimal price would differ from Bravo's because of the different cost structure...but it only makes the example take longer; it doesn't change the conclusion), both companies can make a profit on each widget sold, or so it would seem.

Let's assume that Alpha's factory can make 10,000 widgets per month, and Bravo's can make 15,000 widgets per month. At that optimal price of $4, say that the demand is for 12,000 widgets per month.

Bravo could sell 12,000 widgets at $4 apiece, and make $24,000 per month...if Bravo had a monopoly. Similarly, Alpha could sell 12,000 widgets at $4 apiece, and make $36,000 per month...if Alpha had a monopoly. But neither company has a monopoly. So what happens instead?

The optimal market price drops. Where the new equilibrium is depends on the difference between the cost structures of the two companies, but since we're assuming a constant unit cost here, the calculations become easier.

If the demand were for 10,000 widgets, Alpha would simply undercut Bravo until Bravo exited the market, which would mean Alpha would sell at $2 apiece and make $1 apiece in profit.

But demand is for more than Alpha's capacity. So, let's assume first that Bravo can't do anything else but sell widgets, and Alpha can't increase capacity. In that scenario, the price would rise to $2.01 and Alpha and Bravo would together meet the demand. Alpha would, of course, make far more profit.

If Bravo has other things to make, then the price has to rise to the point where Bravo is willing to make at least 2,000 widgets in order to satisfy the demand...provided that the price isn't so high as to reduce demand back to the 10,000 level (in which case Alpha would sell at just below Bravo's entry price).

Are you following all of this thus far? Moreover, do you believe it to be true? If so, then we can progress further.

 

[Bob Owens]

mweiss,Jan 10 2005, 04:45 PM]

If you're an airline, you probably have contracts with employees that mandate that their wages rise faster than their productivity.

No they "probably dont".The fact is that real wages have declined by around 40% (by the end of our current contract) while productivity over that same period has risen well over 30%.



This was the approach taken by the legacy carriers. Leisure travelers, naturally, have no incentive to look like business travelers, but business travelers have incentive to look like leisure travelers. Once the price differential grew beyond a certain threshold, the cost to businesses of masquerading as leisure travelers became lower than the cost of simply looking like business travelers. It became cheaper to pay for the extra hotel and car rental than to pay for a return on Friday.

So in other words it created higher demand for Hotels and car rentals?

There's a third choice that can be made. Airlines could price based on the lower mode only. They'd sell to the entire demand curve that way. The downside to such an approach is leaving money on the table. Optimum pricing theory would dictate that this is a bad idea. And, in fact, it is a bad idea under most circumstances.

Gee and maybe if McDonalds could figure out how hungry you were they could price their burgers accordingly? However if I paid $3 for my burger and I saw the guy in back of me pay a nickel because he was not all that hungry I think I would get pissed. You may call it optimal pricing but the public sees it as something different. If you practiced this based upon race or gender it would be illegal.


Are you following all of this thus far?

Yes, not only is it basic stuff but its common sense. Keep going.

 

[mweiss]

So in other words it created higher demand for Hotels and car rentals?

It was off-peak stuff, so it didn't have an effect that pushed in the direction of increased capacity. But, yeah, it did shift the hotel and car rental demand curve to the right.

Gee and maybe if McDonalds could figure out how hungry you were they could price their burgers accordingly?

Yes. And, incidentally, the Coca-Cola Company experimented with something very close to that, with vending machines that charged more when it was hot than when it was cold.

However if I paid $3 for my burger and I saw the guy in back of me pay a nickel because he was not all that hungry I think I would get pissed.

Of course. Which is why Coke didn't roll it out after the experiment.

You may call it optimal pricing but the public sees it as something different. If you practiced this based upon race or gender it would be illegal.

It would be illegal because discrimination based on race or gender is illegal. And, incidentally, the public wouldn't see it as something different if the business customer actually got, say, Econ+ seating consistently while the leisure customer got super-tight pitch and no food. The public doesn't get outraged when the same movie theater has a different price for the 2:30 showing versus the 7:45 showing of precisely the same movie in precisely the same theater. Where things got messy in the airline industry was pricing that, while more closely aligned with demand, had the appearence of being arbitrary.

So, let's continue.

In our last chapter, we established that, in a competitive market, the natural market price is at the entry point of Bravo. In a broader sense, if you stack ranked the competing companies by entry price from least expensive to most expensive, the market price would be the highest entry point of the smallest number of companies necessary to serve the market.

So, let's investigate that concept a bit further. Let's look at a market that is reasonably stable. There are three companies (Alpha, Bravo, and Charlie) competing for business, and sufficient demand to profitably support the three businesses. Alpha and Bravo have capacity to support 10,000 units per month, while Charlie has capacity to support 20,000. We'll say that Alpha's entry price is $50, Bravo's is $70, and Charlie's is $100. We'll also say that, at the current stable price of $100 (Charlie's entry point), 30,000 are purchased per month. The three companies each have an equal market share of 33%, and $3,000,000 of revenue would be drawn from this market. Total profit (assuming that the entry points are the same as costs...as before, this assumption just makes the math easier; we'll throw it out soon enough) is $500,000 for Alpha, $300,000 for Bravo, and zero for Charlie, or $800,000 for the entire market.

Now, in comes a newcomer (Delta), who has the same entry point and capacity as Charlie. First of all, a newcomer needs points of differentiation in order to draw customers away from the established businesses. After all, if Delta shows up and charges $100 for the identical item, it will take a huge amount of time to draw customers from the three existing businesses. And, since Delta's entry point is the same as Charlie's, the best long-term convergence point would have Delta gaining 25% of the market.

Of course, Delta might have lower costs than Charlie. If so, Delta can muscle Charlie out of the market entirely...but in doing so must also reduce the stable market price. So, say that Delta's entry point is $90. So, too, would the stable market price (assuming that Alpha and Bravo have lower entry points). The stable quantity demanded would rise (the specific amount would depend on the shape of the demand curve; we'll say that it goes up to 42,000). The market would have $3,780,000 in total revenue, and Charlie would exit the market...assuming that Delta can absorb the entire demand increase (if not, the stable price would remain $100, but market share would be 25% each). This would give Alpha, Bravo, and Delta $400,000, $200,000, and zero, respectively, for a total of $600,000 in market profit.

The market generates greater revenues, but lower profits, after Delta entered.

More to come later. I have real work to do. 😉

 

[mweiss]

Now let's say that Delta is a low-cost entrant. That is, their costs aren't $90; they're $20!

Remember, their capacity is 20,000. So, when we stack rank again, we've got Delta with 20,000, Alpha with 10,000, and Bravo with 10,000. This brings us up to the 40,000 in demand, at what looks like a new stable price of $70 apiece.

But at $70 apiece, the quantity demanded would rise. But to fill that capacity, Charlie would have to re-enter the market. But for Charlie to re-enter the market, the price would have to rise to $100. But if the price rises to $100, the demand falls off below the threshold that would support Charlie's entry...

What happens is that the equilibrium price remains $90, the price at which demand falls off to 40,000 again.

But let's say instead that Bravo is able to produce 20,000 as well. Now we can have the stable $70, assuming that quantity demanded doesn't exceed 50,000 per month. So let's say that it's 50,000 per month...that makes it easy. 🙂

Now we have 20,000, 10,000, and 20,000 sold by Delta, Alpha, and Bravo, respectively. This produces total revenues of $3.5M. Delta, Alpha, and Bravo get $1M, $200,000, and zero profits, respectively. Total industry profits are $1.2M. We can't fairly compare this to the previous $600,000 because total industry capacity changed.

Now, of course, we were holding capacity fixed for each of the companies (i.e., a flat supply curve). That's not very realistic. Instead, what we have in the real world are supply curves that rise to the right (i.e., if you are willing to pay more per item, I'm willing to make more of them).

The math is the same; it's just more complicated. Where above we simply added multiple manufacturers in a stairstep fashion, in the real world you'd add them in a more complex fashion...sort of a rising sawtooth shape. Still, you end up working best where the supply curve and demand curves intersect, which is all that I was describing above.

I need to run now, but next time I'll begin the explanation as to what sorts of market conditions will lead a company to remain serving at a loss.