Chocolate and Gold Coins

Wednesday, August 24, 2005

Innovation Failure

Recently, I have been reading some examples of market failure and I was curious if they might be connected in some way. It seemed to me that they related to an idea I wrote about in this post on magic in the marketplace. Let me list some examples:

  1. I wrote about the fact that all North Indian restaurants in the D.C. area serve a very similar menu. Why don’t we see more variety of recipes?

  2. Coyote writes about how all of the remodeled homes in the Phoenix area are built in exactly the same architectural style.

  3. Half Sigma writes about how there were no smoke-free bars before the big bad government mandated it, even though people like him greatly benefit from having them.

  4. Bryan Caplan wonders why there are so many frame shops in Fairfax County. Why doesn’t the fiftieth frame shop instead become the first custom bookshelf shop?

  5. Until some time in the 1930’s, the only color of automobile you could buy on the U.S. market (and probably other markets) was black.

  6. Baseball famously integrated in 1947 when Branch Rickey chose Jackie Robinson to be on the Brooklyn Dodgers. However, there was a very talented generation of black baseball stars, including the legendary Satchel Paige, who missed out on Major League Baseball. Why did no baseball club before the Dodgers hire Satchel Paige or any other black baseball player?



In all these cases, the free market seemed to fail to produce the kinds of outcomes that we would have hoped. There are certainly specific explanations for each of these cases. But in general, the market can fail to produce certain types of innovations, and it is reasonable to ask, “why?”

Now don’t get me wrong; I’m not claiming that a free market never produces innovations. The free market does produce many wonderful innovations. But when you look at the innovations that really changed our lives, many of them came from spin-offs of military technology, other innovations come from giant corporations that already dominated the market, and many others came from people who never really cashed in on their innovation. Patent law and copyright law does offer some protection for creative people, but many times, the creative person successfully creates a market for someone else to cash in on.

The big money rarely goes to the first entrant into the market, it goes to people like Michael Dell of Dell Computer who came very late to the market and took it over by being a little better than everyone else. Sure, Dell Computer is innovative and I wouldn’t begrudge Dell’s contribution to the industry. But Dell Computer is innovative in a particular way. They aren’t making new goods; they are making existing goods better. This the insight I gave in my article magic in the marketplace: the big money come from people who create technology that no one can figure out how to duplicate.

In each of the above examples, the failure to innovate was primarily because it was so easy to duplicate the innovation. Innovations are public goods: they tell the world that there is a market for this innovation. Free markets don’t produce public goods in abundance. Let’s look at some examples.

In the entire decade of the 1920’s, the only color of car on the market was black. Who would want a black car, especially if everyone else had one? Maybe it isn’t a big deal but it does seem really peculiar. I read somewhere that the black paint was superior to all other car paints at that time. But cannot be the total explanation. Suppose an entrepreneur develops high quality red car paint and tries to sell it to automakers at a slight premium to the original black paint. Would anyone buy it? If Ford wasn’t interested in it, I doubt any other manufacturer would be either. If a small manufacturer buys red paint and tries sell some red cars at a premium; two things will happen: either he finds out that consumers aren’t willing to pay a premium for red cars and he loses money, or they are willing and he has exactly one good year before everyone else in the marketplace copies him. You cannot patent the idea of having a car of a different shade other than black. The innovator has to give away the game to create the market: he has to tell the world that there is a market for his minor innovation.

I go to Indian restaurants frequently and am always struck by the lack of variety of the food. You always get the same thing everywhere. Why doesn’t some restaurant try to be different? Well, suppose that you come out with a wonderful new dish. Can you patent that? It wouldn’t be worth the trouble for a small restaurant to try, even if it were possible. A new dish might make someone a lot of money, but not necessarily the person who creates it. The money might be made by the person who tastes the dish and decides he can make it a little better and a little cheaper.

For many years, all restaurants had smokers and non-smokers mixed together. If you didn’t like smoking, tough, you could eat at home. Restaurants then began offering smoke-free sections, but the smoke would invariably enter that section anyway. And the smoke-free section was typically too small. Bars never made any real effort to make a smoke-free environment. Bar owners believed that their clientele were smokers, but that was only because non-smokers couldn’t stand all the smoke in bars. Why didn’t we see smoke-free restaurants and bars before the government mandated the change? Again, think about the innovator. Either he finds out that there is no market for non-smoking bars and loses money or he finds that there is a market for them and someone else takes it over.

This suggests that there is a role for private charity or government to seed these innovations. A charity that was interested in promoting better public health might have provided some of the capital to start a smoke-free bar. It might have taken only a very small investment to start a revolution. I think the popularity of smoke-free bars shows that it just needed a catalyst to get it going. A charity might have provided a cash reward to the first baseball team to have a black baseball player. Once the color barrier was broken, competition would have done the rest. And charitible prizes (similar to the Nobel Prize) could reward innovators who never really cashed in on thier innovations.

The government could spur innovations by issuing a new kind of mini-patent. I could not justify giving a 17-year monopoly for something like a different color of car paint or a better recipe. But suppose the government issued a 5-year exclusive monopoly for any new innovation not yet on the market. Then we might see a self-cleaning toilet. Surely someone could produce one. The reason we don’t see any self-cleaning toilets is not because they are impossible to make but because they are too easy to make: everyone knows the first mover is going to do all the hard marketing work for somebody else. The mini-patent would prevent any other firm from entering the self-cleaning toilet market for five years. This would give a decent head-start to the first mover. And I guarantee that we will not see a self-cleaning toilet in the next five years so the government would be giving nothing away.

The wonderful thing about market is that, like an oyster, it can turn a small seed into a beautiful pearl. But you do need the seed.

Can anyone think of other examples of innovation failure?

24 Comments:

  • Introducing greater govt regulation to promote innovation !? Seems a bit dubious I must say...

    By Anonymous Anonymous, at 10:51 AM  

  • A very thought provoking post.....reminds me of a class on strategy that focused a lot on disruptive technologies.

    But Dell Computer is innovative in a particular way. They aren’t making new goods; they are making existing goods better.

    The computer industry is actually lousy for comparisons, because say Dell makes a slightly lower end computer to it's competitor, but the difference is so slight, and any way the specifications of the computer are way more than a majority of the customers want (word processing, spread sheets, internet, any computer can handle that) that they can sell more (of course, Dell's innovation was their distribution model independent of showrooms, stores or warehouses).

    Why doesn’t some restaurant try to be different? Well, suppose that you come out with a wonderful new dish.
    In this case, i suspect the problem here is not lack of ability, but aversity to risk. Most indian restaurant owners are (or did not start) well off, and cannot afford to have long months of losses. So they go the tried and tested way....especially in cities where Indian food is not common. But, in some areas, a critical mass of indian restaurants have been achieved. So here, in order to differentiate from a competitor, the restaurant needs to start offering something different. Which is why you see a lot of variety in San Jose, or Edison, NJ.

    The reason we don’t see any self-cleaning toilets is not because they are impossible to make but because they are too easy to make:
    Well put.....but there are some slow changes. For example, here at my university, a majority of the urinals now have been replaced with water-free, self cleaning systems, that actually work surprisingly well. But you are bang on in saying that the first-mover advantage here is almost not there. Still.....given the established players in toilet "technology", market status-quo will be maintained even if they all move to this system.

    The wonderful thing about market is that, like an oyster, it can turn a small seed into a beautiful pearl. But you do need the seed.
    Absolutely, and I think government CAN play a very constructive role in this. This is almost only because only the government will continue to back "basic" research (that doesn't yet have an obvious commercial application). However, 1 out of every 50 or 100 "basic" discoveries ends up having huge commercial applications. This was (and still largely is) one of the greatest strengths of the US, and some thing most developing economies often overlook.

    By Blogger Sunil, at 11:37 AM  

  • Hi Sean and Sunil:
    Sean: I haven't seen you in a while.
    I believe that one of the great value-addeds of government is creating and protecting property rights includind intellectual property. Do you think corporations would ever invent anything without patent law? This is like that.

    Sunil: Thanks for the long thoughtful comment, I really appreciate it.

    I wrote about Dell computer because I was in Austin in 1986-87 when Dell was just starting (if only bought stock) and thought at the time that this college kid starting a computer company after there were already 100 computer companies was crazy. How could he compete with IBM? Somehow, he beat them all. The big money didn't go to the guys who invented the personal computer, it went to people who came to the party late. It often happens that way and that isn't so bad because everyone should be free to improve the technology. But the incentive to innovate is poor because your great idea will make someone a fortune but not you.

    It is curious that Indian restaurants just do not innovate much but it is a sign of a very competetive market ironically. Innovations come from big players in the market. In fast food, McD and Taco Bell always introduce new products. They have incentive because they are big enough players in their markets to capture all the the benefits of their innovations. But if some mom and pop restaurant come out with chapati (which they all know how to make) and it is a hit, then everyone will copy them. You need to have some big players to see some innovative Indian food.

    Thanks for the link to the water-free urinal. That is interesting. The link is hilarious: what is the woman doing there? Do they make urinals for women? How? I have no idea. Or does she just like to look...hmm.

    Anyway, some brave company will try to enter the market for self-cleaning toilets.

    I like basic research too. I think most of the advances in medical care are due to basic research. Private charities like Gates and Hughes are doing good work also. Knowledge is a public good and private firms don't produce much of it unless you pay them to produce it.

    Thanks again for your comment.

    By Blogger Michael Higgins, at 12:41 PM  

  • I am afraid this is a very confusing post. I kind of partly get part of your point, but... I am confused.

    The part I get is that some innovations might take years of basic research to do, and they cannot be patented because they are discoveries rather than inventions... but in that case, do you want to be able to "patent" discoveries even for a short period? Does that even make sense. On the other hand, if they are genuinely patentable and if they have taken lots of time and effort to develop, then the argument should be that they should be fully patentable, perhaps for more than 17 years.

    I also kind of get the idea that some innovations are so easy to make that they deserve a shorter patent than a 17 year one. I think that most software patents would come under that. But I would support those mini-patents primarily to avoid giving longer patents rather than to reward currently unpatentable ideas.

    What I don't understand is your arguments about new recipes and smoke-free bars. Surely it doesn't take years of research to introduce a new recipe in a restaurant? You just have to cook it, change the menu, and do some publicity. How long can the payback period be? Why would patent protection be required for that? If you develop a reputation for continuously introducing innovative recipes, why would someone go to a copycat competitor, even if he offers it at a cheaper price?

    Similarly, for smoke-free bars, the answer is more likely to be advertising and creating a brand image, so that your bar is indelibly associated with "non-smoking bar" in a non-smoker's mind. If you manage that, then competitors will have a tough time catching up. In any case are you proposing that someone should be able to ban others from opening non-smoking restaurants on the ground that he thought of it first?

    Finally, about the black vs. red car, I am puzzled about that too.

    If the problem is R&D expenses in developing red paint, then surely the technique to develop red paint can be patented?

    If the problem is large fixed costs in setting up a red-car plant, then surely that by itself should act as an entry barrier for others? What is the need for a patent? Won't a patent granted to a car manufacturer for a new idea actually hurt the paint manufacturer in this case, because he will be forced to sell his paint only to the car guy?

    If developing and manufacturing red paint is trivial, the only problem is market acceptance, then surely it should be possible to test-market it with little risk of failure? Shouldn't the rewards be proportionately less, because the costs involved are less?

    So I guess I am saying that I understand your concept, but none of your examples seem to be a good illustration of the problem...

    By Anonymous Anonymous, at 2:15 PM  

  • Hi Michael,

    Nice blog there.


    This suggests that there is a role for private charity or government to seed these innovations. A charity that was interested in promoting better public health might have provided some of the capital to start a smoke-free bar.

    Isn't that the role of venture capitalists?

    Although I agree about the intellectual property rights, as an Indian, I reserve the right to totally distrust the governments :-). Whatever laws they enact would really only benefit the politicians and maybe a few companies that have some kind of pact with the ruling party.

    As far as India is concerned, I feel more venture capitalistm, and more judicious lending by banks would go a long way in finding new talent and ideas.

    BTW, did you catch this bit of news

    -swami

    By Anonymous Anonymous, at 3:40 PM  

  • Hi Ravikiran
    I appreciate your long comment.

    I think part of the confusion here is that I am trying to tie together many different examples into a nice package when they really are different. I wouldn't deny that they these examples might require different responses (which I talked a bit about) and it may be that in some cases the best response would have been to do nothing. After all, in the end baseball did integrate (although too late for Satchel Paige) and cars did eventually come in a vareity of colors and I can always go to a Chinese or Italian restaurant if I get tired of Indian food.

    I think you're right that this is confusing because I myself don't really understand things like why Indian restaurants all offer the same cuisine. They don't have to spend money to find new recipes, they can just read Tarla Dalal or someone else. And they know how to make chapati, but no one does it. It might take a little capital to mix up chapati dough, and maybe there is a risk there, but it seems small. But the key here I think is it doesn't pay for small mom and pop shops to innovate: they don't get enough of the market to benefit from the innovation. If chapati caught on, every Indian restaurant would have it in no time, because all that is necessary is to find out what is the price customers would be willing to pay for chapati. But I don't see any kind of patent justified here. It needs a minor catalyst: someone who is willing to make a minor risk.

    The red paint vs. black paint is curious also. It is bizarre to think that all automobiles were black for twenty or thirty years. I don't claim to understand this very well but I would be willing to bet that there was essentially no way to patent colored car paint. Also, there was a famous quote associated with Henry Ford that went something like, "Customers could have cars in any color they liked as long as it was black." He obviously didn't care too much about the color and figured the customer couldn't care less also. If the market leader doesn't innovate, the followers probably won't also because the market leader will always be quick to capitalize on any successful innovations.

    I disagree that the market testing would have been inexpensive. Market testing could be very expensive. If you just poll people, it might not cost much but can you really trust a respondant who says he would pay an extra $100 for a red car? You would have to make them and sell them to see what the market is like and if it is a hit, you only own the market for one year. If it is a failure, you are stuck with repainting the cars.

    The smoke-free bar was curious because I think a lot of non-smokers wondered why no one created them. I'm not too familiar with the bar scene but I recall that all of the bars that I happened upon had just a disgusting stale cigarette smell about them so I can believe that the market produced too few (probably none at all) smoke-free bars. Why is this? Why wouldn't someone risk producing a smoke-free bar and advertising it. It might have been that someone tried it, was otherwise incompetent as a businessman and went out-of-business. This would have convince other bar owners that the smoke-free bar was a bad idea.

    But I think the real reason why we didn't see the smoke free bar was because it would have taken quite a bit of marketing to get one going. That would have been a considerable risk for something that could not be patented or copyrighted. It needed a catalyst.

    The mini-patent might have worked there if the patent owner could have had the legal right to license his "smoke-free" bar to other bars. After the one bar make a success of creating a smoke-free bar, others would want to copy him, and might have paid $10000 or more per year to have that right, maybe.

    Yes, there might be more important examples of innovation failure, but these I was more certain about, even if they were trivial. The question is, do you need the government to mandate better fuel economy? Does the free market produce more fuel-efficient cars or does the government have to regulate, and if so, how should they regulate? The U.S. does regulate in this way. Is it a good thing or a dumb thing? I'm not sure.

    The thing I don't like about patents is that they are almost useless to protect innovators. They only protect your invention, but do nothing to protect your investment in creating a market that never existed. You do all the hard marketing work to show the world that there's a market for your widget and that market is taken over by someone else who makes a similar product that skates around the patent protection.

    But patents are always a bit controversial because they are monopolies and after the invention is invented people want to see everyone have the right to enter the market. This ignores the fact that the market might not have existed without the original patent protection.

    Btw, and I hope you don't this wrong way, but I sure hope you post something new on your blog soon. I keep going there and its the same post.

    Thank you for your comments.

    By Blogger Michael Higgins, at 4:08 PM  

  • Hi Swami
    Thank you for the link, that is a cool story (about a Vietnamese doctor who made his own homemade endoscope).

    I agree that venture capital is the way to go for most new innovations. But a venture capitalist wouldn't be interested in funding a smoke-free bar because you cannot patent that idea.

    I cannot disagree that new patent laws in the hands of an incompetent and maybe corrupt government could do more harm than good.

    By Blogger Michael Higgins, at 4:27 PM  

  • Wow Ravikiran! I agree with you wholeheartedly. The Sun will rise in the west tomorrow.

    Patent law has gone too far and has become an obstacle for innovation. That is what one research study said. Patent/copyright laws are a kind of agreement between various interests in the society. Lastly, a lot innovators/inventors don't do it for the money but the thrill of it.

    Apparently market is not God. Who could have guessed?

    sean, it is not introducing greater regulation but introducing public money into private coffers. I am all for it as long as the whole society benefits (say computers, internet, subsidy for expansion of telephone systems in the US, postal service etc) but certainly against it when it is simply a transfer of resource from the masses to a few private hands.

    yum yum

    By Anonymous Anonymous, at 9:29 PM  

  • If someone tastes the dish and decides he can make it a little better and a little cheaper (for instance), isn't that good for the public? Why should we make the guy pay a penalty for that? The original innovator is free to try to improvise anyway.

    Innovators and those who come up with original ideas should not be discouraged by the fact that others make money out of their work. As you rightly pointed out there should be other kind of rewards (like prizes) for the innovators. Coming up with innovative ideas is rewarding in itself too!

    By Anonymous Anonymous, at 12:59 AM  

  • Michael, thanks for your response. About smoke-free bars/restaurants, I have a doubt. What was impossible to find before the ban. Was it smoke-free restaurants or smoke-free bars? It occurs to me that there is a subtle difference between the two cases in the social expectations that could lead to different outcomes. In a restaurant, you go there for the food, the quiet atmosphere and the ambience. It is also likely that you go there with your family. Even if you go with a group of friends that is a mix of smokers and non-smokers, it is acceptable for non-smokers to insist that everyone at the table not smoke, so it is easier to ask for a non-smoking table or section.

    On the other hand, at a bar, people typically go out to have a good time, usually with a group of friends. So if you have a mixed group, the pressure is usually the other way, i.e. on non-smokers to acquiese to their friends smoking, because after all, they are out to "have a good time" and it would be churlish to ask people not to light up. So this translated to asking for smoking tables and smoking sections. So non-smoking bars would not really become popular.

    My understanding is that this has been changing in the US, because militant non-smokingism is becoming more socially acceptable. The government ban short-circuited the process.

    By Anonymous Anonymous, at 1:01 AM  

  • Yeah I need to update my blog sometime soon...

    By Anonymous Anonymous, at 1:02 AM  

  • This comment has been removed by a blog administrator.

    By Anonymous Anonymous, at 1:04 AM  

  • This comment has been removed by a blog administrator.

    By Anonymous Anonymous, at 1:27 AM  

  • My point about the paint was that if it were a new technology, it would be patentable under the then existing laws. I wasn't referring to the "idea of having a red coloured car", which would be patentable only by Amazon's one-click shopping standards. (Now that would be a good candidate for the mini-patent.)

    My understanding is that the problem with different coloured cars was not the colour, but the cost of changing the assembly line to be able to paint cars in different colours. If that was the case, then anyone who invested in such an assembly line would have a first mover advantage, the advantage of an entry barrier because others too would have to invest in the changes themselves, and a continuous advantage because he would be benefiting from economies of scale which he got by grabbing a higher market share because he moved first.

    My point is, in this case the idea is not such a great one that it ought to be rewarded with a patent on top of the advantages he already has. I am sure others got the same idea of different coloured cars too.

    Hey by the way, I think you are puzzled by those long lead-times of 20-30 years because you are too young to remember the times when business lead times were really that long. You've been spoilt by the internet age which goes through 10 cycles in one year ;)
    (Yeah, I am younger than you, but I like playing an older and wiser man on the internet.:) )

    By Anonymous Anonymous, at 1:33 AM  

  • This comment has been removed by a blog administrator.

    By Anonymous Anonymous, at 1:36 AM  

  • Um... Sorry for hogging the comments section of this blog, but it occurred to me that it might be interesting to study the evolution of Indian food in the US with Mexican food in the US..

    Specifically, if we can answer
    1) When was Mexican food introduced?
    2) Were the restaurants started by immigrants?
    3) Were they first patronised by immigrants?
    4) When did restaurant chains like Taco Bell start? Who started them? First-gen immigrants? Second-gen immigrants? Non-Mexicans who wanted to cash in on the popularity of Mexican food?
    5) Who are the ones innovating there? Small restaurants? Big chains? Or the former introduce and the latter popularise?
    6) What sort of innovations? Did they just get recipes over from Mexico or did they come up with innovations uniquely suited to Americans' palate?

    A comparison of the trajectories the two cuisines are taking would tell us a lot about the cultural differences between the two ethnicities. We Indians may also learn something about ourselves...

    Now I will get back to work..

    By Anonymous Anonymous, at 1:50 AM  

  • Hi Michael, my point was the process of introducing another category say mini-patents would involve a whole process of subjective decision-making on part of the state apparatus - something which has not worked well going by the past. For e.g. an inventor may invariably prefer a longer patent period - but the concerned state apparatus may deem it fit only for 5 years (mini-patent). The same patent may spark further developments in the field, in which case our poor inventor will lose out on his deserving royalties. This is what I meant when I said greater govt involvement seems a bit dubious in this case. In my book this does not pass the justifications for greater governmental action - seems more like a case for increasing red tapism and stifling innovation.

    By Anonymous Anonymous, at 2:06 AM  

  • Hi Ravikiran
    You're certainly welcome to post as much as you like in my comment section. Your comments are excellent. I think I may need to respond to one or more of your comments with another post - especially the one about Mexican food.

    By Blogger Michael Higgins, at 7:52 AM  

  • Hi Yum Yum, Anand, and Sean
    Yum Yum: I think patents can work sometimes for some products. It is important that innovators have their intellectual property protected. The length of the patent guarantee is kind of arbitrary thought and I'm not sure how long it should be but I think it is completely ridiculous to pay a royalty on song written 100 years ago. How much of an incentive to innovate is there for someone if he knows he might get a royalty from a song 100 years from now?

    Anand: sure, innovating on others inventions makes for good cuisine just as writing a post based on another's post makes for good blogging. The only issue there is that the original innovator might not get much return on his investment of marketing a new product. But in the case of food, it doesn't seem like a really big deal. I'm not suggesting some kind of government intervention on encouraging recipes or something.

    I think prizes and other kinds of plublicity might help spur competition in restuarant and in other areas as well.

    Sean: I agree that the mini-patent might cause some political problems. How long should they be. Do they go to the politicians buddies? How does it work? The mini-patent is just a "here's a thought" kind of idea that needs more refinement for sure.

    By Blogger Michael Higgins, at 3:05 PM  

  • Hi Ravikiran
    I agree about the difference between the restaurant (more family oriented) and the bar (not family oriented). Yes, the fact that you take your kiddies to the restaurant will make parents want a smoke-free environment. It took a while for that to happen.

    It is true that non-smokers' tolerance of smokers has gone down with time. And as non-smokers are less likely to encounter cigarette smoke, they are less likely to tolerate it when they do. So I doubt most non-smokers would be willing to hang out with their smoker buddies in the bar anyway. A lot of non-smokers today just can't stand the smell.

    To me, although I don't like smoking and can't stand the smell, it seems a pity that smokers wound up with no bars to hang out in (at least in New York). The market should have split it down the middle and given half smoking and half non-smoking, but it didn't happen. I'm not really sure why.

    By Blogger Michael Higgins, at 3:13 PM  

  • Hi Michael
    Xerox is one such example of failed innovaters, I think. Xerox and Xerox park has most innovations at one point in time, held their patents, but over a period of time lost out to better marketers. Just like IBM did with Microsoft though I feel that OS2 was a better and more robust product than Windows.
    Sourin

    By Blogger Sourin Rao, at 8:26 PM  

  • I have some thoughts, but until I organize them ... you were in Austin in 1986-87? So was I (1984-92, actually). Why didn't we meet and get in a game of tennis at the UT courts?

    By Blogger Dilip D'Souza, at 4:59 AM  

  • Sorry to burst your bubble but slef-cleaning toilets have been on the market for a while

    According to a June 2004 issue of Nature, "a quarter of all toilet bowls coming on to the market in Taiwan …now come with self-cleaning nano-coatings."

    see http://smarteconomy.typepad.com

    By Anonymous Anonymous, at 12:46 PM  

  • Great post.
    This post is very meaningful.
    Thanks for sharing this post.
    Keep posting

    By Anonymous Cleaning services in London, at 2:44 AM  

Post a Comment

<< Home