Tuesday, June 18, 2013

Could someone explain the market failure that protecting car dealerships solves?

The Wall Street Journal has a good story today about how car dealerships are (successfully) lobbying legislatures to ban Tesla Motors from marketing their cars directly to consumers.  GOP legislators, who get the willies about regulation that actually solves real problems, are on board with supporting protectionist policies for auto dealerships.

Does anyone really think that the industrial organization of the automobile retail industry works well?  My family buys a car every five years or so, and our experience is that no one tries to exploit asymmetric information like auto dealers.  I have lots of reasons to believe that our experiences are not unique.

What amazes me is that even in the age of the internet, when one can use sites like Edmunds to figure out what to pay for a car, dealers start out by assuming that the consumer is stupid, hope they get an absurdly marked up price, and only get reasonable when they find out their customer actually knows something.

Elon Musk is a visionary in many ways.  With the Tesla, he might make two important contributions--he might free  from petroleum, and he might free us from car dealers.




13 comments:

David Ruggles said...

Where to start on this one. I'll have to resond to this one in steps. The first thing that needs to be understood is that so called "dealer protection" entails protecting auto dealers from their own suppliers, the auto manufacturers. Those same manufacturers understand that trying to sell directly to the public is a losing proposition. First, they don't have the money to become their own dealer network. Second, experiments with manufacturers owning their own outlets have failed miserable with every attempt. Saturn and the Ford Collection would be examples.

Let's label Elon Musk a visionary AFTER he has made his idea work, not before. Make sense?

David Ruggles said...

http://autosandeconomics.blogspot.com/

Michael Smitka said...

Just a quick pointer toward the Autos & Economics blog that I do with David Ruggles, who's spent a lifetime on the distribution end of the industry.

On the legal side, tensions are inherent in franchise systems, on the margin a franchiser always wants more franchises, while a franchisee doesn't want an additional (here) dealership just down the road. Add in cyclicality, and that manufacturers financed their operations with the capital of dealerships (cars are paid for when they leave the assembly line, so in normal times manufacturers hold very little finished inventory) and these tensions are amplified. Franchise law developed to provide a set of rules to put bounds on disputes as a lower-cost alternative to lawsuits. Of course that doesn't mean that the rules, which vary state-by-state, are sensible – at the state level dealerships have (political) clout, manufacturers typically do not.

As to asymmetric information, there are many components. In the background however is that cars are highly differentiated (though not for most brands fully customized) so that comparing pricing is in fact hard as you move away from base models. Equally, most transactions are not a straight purchase, but involve a tradein (where transaction costs are high and asymmetric information higher), finance, and (frequently) insurance (extended warranties).

Finally, depending on the brand/franchise, there are "stair-step" incentives ["trunk" money] that mean a (retroactive) volume discount when sales hit a certain number, on and on. This means that there really can be end-of-month bargains, and there are model-year effects and on and on. This means that there is heterogeneity of costs among dealerships that are hard to search. In addition, supply and demand change.

Alongside real estate cars do remain one of the few items over which consumers haggle over prices.

More on my blog...

David Ruggles said...

RE: Does anyone really think that the industrial organization of the automobile retail industry works well?

Ruggles: YES!

RE: "My family buys a car every five years or so, and our experience is that no one tries to exploit asymmetric information like auto dealers. I have lots of reasons to believe that our experiences are not unique."

Ruggles: Your opinion. What part of the process do you not like? Does it trouble you that the auto dealer tries to make a profit? Do you not like the opportunity to bargain your way to a better deal than someone else received?

RE: "What amazes me is that even in the age of the internet, when one can use sites like Edmunds to figure out what to pay for a car, dealers start out by assuming that the consumer is stupid, hope they get an absurdly marked up price, and only get reasonable when they find out their customer actually knows something."

Ruggles: So do you want to arrange it so all consumers pay the same? Dealers are discouraged from price fixing by something known as "jail time." Dealers need to make around $3K per transaction. If YOU get a deal for $2K, the next guy needs to pay $4K. Dealers would be fine with someone coming along to force everyone to pay the $3K. But consumers wouldn't be happy with that either, would they?!

Anonymous said...

I have some suspicion that the sites like Edmunds are also subverted by the auto complex so that the numbers we see there are also inflated.

JT said...

@David Ruggles - what value are dealers adding? And more specifically, what value is being added to our society by not allowing Tesla to sell direct to consumers?

Clearly there is value to dealers, and to politicians they buy. But on aggregate, is it a net gain to society?

derrida derider said...

Messrs Smitka and Ruggles, I'm confused. You argue that the car dealership approach serves consumers well. You may be right as you clearly know a lot more about this than others here (though I might just note my experience as a consumer is similar to Richard's).

But how on earth is that a case for legislatively forbidding other approaches? Maybe consumers do prefer dealerships, with their non-transparent costs and conflicting set of incentives (that it is hard to compare cars as you move away from base models, BTW, is deliberate strategy to increase the asymmetry of information). After all, there will indeed be substantial assurance costs with direct sales. But that is a matter for the market, not captive legislators, to settle.

As Richard says, show us the market failure here.

Michael Smitka said...

1. Since Tesla has no dealerships, I see no reason why they should be forced to create them. The arguments Ruggles & I provided above for why dealership law is irrelevant – there are no independent dealers who need to be protected from their monopoly OEM supplier. I believe Ruggles concurs in that.

But to step back and rephrase as an economist, it's sensible to have legislation to protect franchisees from franchisers (of which the auto industry is but one example). This is (among other things) the standard hold-up problem, as franchisees are asked to invest in specific assets.

I see no reason why a manufacturer or wholesaler should be forced to use a franchise system for consumer sales. Mixed systems are the problem.

2. Will Tesla be able to make its strategy work, legal issues aside? I think that its strategy has worked so far because it is focused on a high-end niche market and so (among other things) need not provide the ability to take a trade-in. I wrote about that on the Ruggles/Smitka blog Here.

3. Economies of scale have increased (or at least are capable of being realized). In the early days of the industry (the 1920s) every small town had a couple dealerships. Over the past 50-60 years the very success of the auto industry led to better roads, while other changes led urban systems to expand. The potential market area is much larger, as small towns are now well-connected suburbs. This increased rivalry among dealers, compressing margins and encouraging exit. The flip side is that today car dealerships make very little money on new cars -- that didn't used to be the case.

It would be neat to get input from an urban economics perspective on this – I've done a bit in that area, but have not tried to look at auto distribution in any formal way, co-location [urban areas have 2-3 "car alleys" while the old downtown dealerships are almost all gone], measures of density and dispersion, and whether there's a convergence between dealers on one or another metric. For example, Toyota was a late entrant and so has a low density compared in particular to Ford and GM. But some of that's an artefact of including lots and lots of small rural dealerships in the averages, and Toyota has virtually none. So do Toyota dealerships and urban Ford ones look more similar over time?

4. I was very fortunate to have 2 dealerships share their financials with my students and I this past spring (Ruggles helped us go through one set of them). It's a fascinating business, with a mix of profits from new cars (sometimes negative), used, service, finance and insurance (extended warranties) and (at some dealers) a body shop.

A trend I've not examined is the growth of multi-franchise dealers, the top 75 or so dealers all have revenue over US$1 billion. However, overall the top 100 dealers account for a very small share of industry sales, the Herfindahl index is effectively zero.

David Ruggles said...

As I might have mentioned, the average dealership needs to average $3K per transaction. This can be a combination of vehicle sales gross profit and F&I gross profit. In some cases, the front end gross might be $1K with $2K on the back, or any combination that reaches the average. A Dealer can't take $1K deals without having the opportunity to make $5K to make the average. These figures are approximate figuring an average vehicle cost of $30K. High line stores need to average more while stores with a lower average vehicle cost might average less.

$3K is about 10%. Unreasonable? Perhaps we should mandate a $3K profit so Consumers will be happy? After all, they don't want to pay more than anyone else, and that's what they complain about with the current system which is based on negotiation.

David Ruggles said...

RE: "I have some suspicion that the sites like Edmunds are also subverted by the auto complex so that the numbers we see there are also inflated."

Are you surprised there is a profit motive? Who do you suppose pays for those facilities, the insurance, sales staff, support staff, inventory costs, insurance, etc.? How much is left after the Dealer makes his/her $3K? Consumers don't care about that.

RE: "What value are dealers adding?"

Immense. Read the Maryann Keller entries on autosandeconomics. Again, facilities, inventory buffer, sales staff, support staff, etc. Someone has to pay for that.

RE: "And more specifically, what value is being added to our society by not allowing Tesla to sell direct to consumers?"

Professor Smitka and I seem to be of the same mind. Neither of us has a problem with Tesla owning its own dealerships as long as there are no privately owned franchisees. As Mike so aptly put it, "The problem is with a mixed system." So now I will tell you what will happen. Tesla should be allowed to own its entire dealer network. As soon as Elon Musk figures out he is hampered by that system, he will sell his private network to private dealers so his expansion can continue.... that is if his sales and production actually support profits ongoing. IF he has been successful, his private network will have great value. I suspect that is his calculus.

Clearly there is value to dealers, and to politicians they buy. But on aggregate, is it a net gain to society?

David Ruggles said...

As a practical matter, when a capitalist makes an investment, the first objective is not altruism. That investment won't be made if that capitalist has to compete with his/her own supplier. IF Dealers are eliminated, which is highly unlikely, and OEMs own their own sales networks, they are free to price where ever they want. Price Fixing? Is that what Consumers want?

David Ruggles said...

RE: "Maybe consumers do prefer dealerships, with their non-transparent costs and conflicting set of incentives (that it is hard to compare cars as you move away from base models, BTW, is deliberate strategy to increase the asymmetry of information)."

Yes, there is a deliberate strategy on the part of car dealers to withhold their cost information. Kindly explain to me why that information should be privy to Consumers in the first place. Do you get transparency at the grocery store? The clothing store? WalMart? Furniture? Jewelry? Why would anyone feel entitled to complete cost transparency at the auto dealership? If you want to know pure costs there is a way to find that out. Buy a dealership. MAke the investment. Hire all the employees. Sign a long term lease. Sign capital loans personally, so that if the business fails, your house and personal wealth is the first to go. THEN you can see all the bare cost figures up close and personal. Until then, it is none of your damn business.

Glad I'm not an owner any more so I can express myself!

David Ruggles said...

Correction: I did not mean to say that Saturn was an example of OEMs owning their own sales outlets but it is an example of an OEM effort to establish a dealer network outside of the traditional network that has been proven to work. I should have made that clear.