Today’s New York Times talks about cases where Wikipedia showed a death before the police or newspapers knew. I am quoted: Robin Hanson … said … these examples are hardly evidence of predicting the future. Rather, he suggested, it was "a bit newspaper-centric to say that news has not broke ‘publicly’ if it is being discussed online in rumors but has not appeared in a newspaper." He added that "with more and more kinds of media, there are more and more intermediate levels of info availability."
I took Robin to mean that at some point the costs of incoherence will outweigh the benefit of ease of enforcement, not that a simple system is always better. Fit between a rule and the underlying policy goals (substantive accuracy) and ease of enforcement ('clarity') are both important. Good rules craft some reasonable compromise, though there is no single perfect balance for all purposes and a particular rule may lie at one end or the other of the spectrum. A rule may be clear but a relatively poor fit with the underlying policy goals. The novelty requirement in patent law is an example. Or a rule may be an excellent fit with the underlying policy goals, and yet very unclear and uncertain in its application. The non-obviousness requirement in patent law is an excellent example. Both inaccuracy and lack of clarity may be defensible in particular circumstances. I took Stuart's original point to be that a rule may appear to be 'incoherent' (indefensible) if considered solely from the substantive accuracy perspective, and yet can be quite coherent, or at least a reasonable compromise, from the enforcement perspective. I think this is an important point which is often overlooked. For example, my own view is that both the idea/expression dichotomy in copyright law and the policy/operational distinction in the law of public authority liability, and primarily driven by evidentiary (enforcement) considerations, but they are invariably assessed from a substantive perspective.
To return to the topic of this blog, overcoming bias, there is a third category of rules that are neither accurate nor easy to enforce. I would call such rules truly incoherent, as opposed to inaccurate or unclear. Why would truly incoherent rules ever arise? I think it is very plausible that cognitive biases would result in a natural category such as 'public' being systematically misapplied because of a psychologically salient resemblance to some aspect of a legal problem. I take it that that this is what Robin was originally getting at. This is a very interesting point, but I just don't think that it is illustrated by the use of 'public' in patent law.
Stuart and Norman, I agree that ease of enforcement should be an issue, but surely there should be limits to how incoherent a concept we will use to make enforcement easy.
Why? I like your solution for insider trading (once we solve the friend/proxy issue - Byrne's idea seems good for information, but unenforceable in the criminal sense), but I don't agree with your statement here.
An incoherent system has costs - selective enforcement, exploitation by special interest, requirements for complicated legal knowledge, loss of respect for the law. But if the issue it is addressing is severe enough, then there may be a basis for the incoherent system (such as, for example, the British official secrets act). It's a fallacy to believe that a simpler system must be better than an incoherent one.Life is incoherent. Death is simple.
Robin, I entirely agree that there are limits to how incoherent a concept we should be willing to use to facilitate enforcement. But I think that the concept of 'public' is really quite coherent in the patent law context in particular. It is relevant to the novelty issue, where it asks in effect whether anyone else had the information in question without legal restriction on their ability to use it. There don't seem to be particular difficulties in its application and I think it fits reasonably well with the underlying policy goals. (The non-obviousness criteria, on the other hand, is extremely difficult to apply in a manner that is consistent with the underlying policy considerations -- but that is not because of any difficult in the concept of 'public' which is only tangentially relevant.)
That is not to say that the concept of 'public' is coherent in the insider trading context. I don't know enough about that area to have my own opinion, but I see no reason why it need have the same content in both contexts.
Stuart and Norman, I agree that ease of enforcement should be an issue, but surely there should be limits to how incoherent a concept we will use to make enforcement easy.
Re the patent example, I'd second Stuart Armstrong's point that the distinctions are based largely on ease of enforcement, and more particularly evidentiary concerns. The basic question in patent validity is whether the invention required the lure of a patent to bring it forth. If anyone made it public without bothering to apply for a patent, this is good evidence that a patent wasn't really required. Of course, it's not conclusive evidence as the disclosure might just have been inadvertent, but you have to draw the line somewhere.
Incidentally, while the US has a one-year grace period for disclosure by the inventor (you can get get a patent if the invention was disclosed by you in the year prior to the application date, but not if it was disclosed by someone else), European countries generally do not. I believe that the Cohen / Boyer patent, which is the basic patent on GM technology, was granted in the US but not in Europe because they had published an academic article disclosing the invention prior to applying for the European patent. Another interesting example is that the windsurfer was not patented in the UK because a 12 year-old boy made an used one for a summer at the family cottage in 1958: see Windsurfing International Inc v Tabur Marine (Great Britain) Ltd [1985] RPC 59(CA).
An invention might never have been disclosed and yet be obvious, in the sense of not requiring the lure of a patent for its development. The classic example is the automobile: it was obvious to mount an engine on a cart as soon as the engines became light enough. Such an invention is not patentable in principle as being obvious, so it is not really right to say that "you can keep patent rights even if lots of other people said they already thought of your obvious idea." Non-obviousness is indeed proven by testimony of experts in the field that the invention would have been obvious to a person with ordinary skill in the art. However, non-obviousness is considerably more difficult to establish than lack of novelty, as the courts are very aware that a truly worthy invention may appear obvious with the benefit of hindsight. And indeed, a Mr. Selden did get a US patent on the automobile, to his considerable profit.
Re insider trading, the Journal of Legal Studies Vol 9(4) (1980) is a special issue on the law and economics of privacy, and includes a couple of papers on insider trading.
"If insiders are required to preannounce trades, doesn't this make the info about preannouncement also insider information & allow for explotation?"
Of course the "preannouncement" information becomes insider information, but how do you suggest it could be exploited? It would be impossible for this new insider--the one with inside preannouncement information--to trade based on the new information unless he preannounces his intention.
Stuart: if you wanted such a fund to be successful, you could test how much insider information was dispersed by looking at the stock's performance after earnings surprises. In general, the way to measure how much insider information spreads is to see how the stock reacts to newly-public data. If an executive claims to be registering his trades with the fund, but the stock drifts upwards the month before a good earnings announcement, you could presume some amount of chicanery.
Fergus, great question. Husband of a lawyer? 'One financial expert'? Robin's reported source and linked confirmation gives me the sense this may be a protean urban myth.
"Legal definitions of insider trading have expanded beyond all common sense in recent years. One financial expert told me not long ago that I would be guilty of it if I saw a factory on fire from my plane as it landed in Des Moines, inquired about it, learned that the factory was owned by a company in which I had stock, and promptly telephoned my broker to sell before the news reached Wall Street."
Other alternative: what about funds, tracking exactly the purchases of insiders (and no restrictions on what insiders can buy). That way, wary investors could invest in these funds as a sort of "insurance against insider dealing", and academics could finally get an idea of what the information advantage on insiders really is.
Still not sure how to deal with proxy buying and passing info on to friends, though.
It seems "public" vs "private" in laws are based on mainly on ease of enforcement, rather than on coherent concepts.
The idea of "classifying traders into various levels of how informed they are" is a much better concept, but may suffer from enforcement problems. The major one concerns the children, friends and relatives of the "highly informed" traders. Traders may pass on information to them or use them as proxies. How to deal with this issue, without resurrecting some concept of "public" and "private" information?
On a side note, insider trading laws (just by their existence, not by the fact they do much) go a long way to improving the image of financial capitalism in general, and seem to encourage people to invest (especially those "overconfident investors" who contribute a lot to making the markets informative). Tinkering should be done with care, lest these advantages be lost.
BTW, did a search on Google, and haven't found any primary source for that factory fire story - it would be good to know more of the details, as it seems very pertinent to the issue. But it sounds like it might be an urban legend.
Fergus, I heard that one from the husband of a lawyer. Did I get it wrong? One source agrees with me.
Tuukka, yes info can often be "roughly" classified, but rough classifications do not support sharp legal distinctions well. I don't understand your concern about preannouncement.
There is information within organizations (firms, gov. bodies, etc.) that can be roughly classified as confidential or public. My understanding is that insider information is confidential information. That is, access to it is limited to some group by a contract or a law. Many ppl argue normatively that inverstors in the markets should have equal ground regarding the access to information. It doesn't seem good idea to trade with people with insider information, since on the average (& other things equal) you are losing.
When you are classifying traders in to various categories according to how informed they are, are't you implicitly making a categorization with respect to information as well?
If insiders are required to preannounce trades, doesn't this make the info about preannouncement also insider information & allow for explotation?
Maybe the solution could be "private law" or just firm-specific contracts which regulate insider trading. The solution might be something near present situation. To make the stock of company attractive for traders you might want to try exclude insiders from the market (also: some kind of insider regulation removes the incentive for bad management decisions by managers associated with gains in the stock market)
When and where did "insider trading laws punish ... a man who traded on the "private" info of seeing a factory fire when looking out an airplane window" ?
The patent law may concern more about written / published, rather than private or public.
I took Robin to mean that at some point the costs of incoherence will outweigh the benefit of ease of enforcement, not that a simple system is always better. Fit between a rule and the underlying policy goals (substantive accuracy) and ease of enforcement ('clarity') are both important. Good rules craft some reasonable compromise, though there is no single perfect balance for all purposes and a particular rule may lie at one end or the other of the spectrum. A rule may be clear but a relatively poor fit with the underlying policy goals. The novelty requirement in patent law is an example. Or a rule may be an excellent fit with the underlying policy goals, and yet very unclear and uncertain in its application. The non-obviousness requirement in patent law is an excellent example. Both inaccuracy and lack of clarity may be defensible in particular circumstances. I took Stuart's original point to be that a rule may appear to be 'incoherent' (indefensible) if considered solely from the substantive accuracy perspective, and yet can be quite coherent, or at least a reasonable compromise, from the enforcement perspective. I think this is an important point which is often overlooked. For example, my own view is that both the idea/expression dichotomy in copyright law and the policy/operational distinction in the law of public authority liability, and primarily driven by evidentiary (enforcement) considerations, but they are invariably assessed from a substantive perspective.
To return to the topic of this blog, overcoming bias, there is a third category of rules that are neither accurate nor easy to enforce. I would call such rules truly incoherent, as opposed to inaccurate or unclear. Why would truly incoherent rules ever arise? I think it is very plausible that cognitive biases would result in a natural category such as 'public' being systematically misapplied because of a psychologically salient resemblance to some aspect of a legal problem. I take it that that this is what Robin was originally getting at. This is a very interesting point, but I just don't think that it is illustrated by the use of 'public' in patent law.
Stuart and Norman, I agree that ease of enforcement should be an issue, but surely there should be limits to how incoherent a concept we will use to make enforcement easy.
Why? I like your solution for insider trading (once we solve the friend/proxy issue - Byrne's idea seems good for information, but unenforceable in the criminal sense), but I don't agree with your statement here.
An incoherent system has costs - selective enforcement, exploitation by special interest, requirements for complicated legal knowledge, loss of respect for the law. But if the issue it is addressing is severe enough, then there may be a basis for the incoherent system (such as, for example, the British official secrets act). It's a fallacy to believe that a simpler system must be better than an incoherent one.Life is incoherent. Death is simple.
Robin, I entirely agree that there are limits to how incoherent a concept we should be willing to use to facilitate enforcement. But I think that the concept of 'public' is really quite coherent in the patent law context in particular. It is relevant to the novelty issue, where it asks in effect whether anyone else had the information in question without legal restriction on their ability to use it. There don't seem to be particular difficulties in its application and I think it fits reasonably well with the underlying policy goals. (The non-obviousness criteria, on the other hand, is extremely difficult to apply in a manner that is consistent with the underlying policy considerations -- but that is not because of any difficult in the concept of 'public' which is only tangentially relevant.)
That is not to say that the concept of 'public' is coherent in the insider trading context. I don't know enough about that area to have my own opinion, but I see no reason why it need have the same content in both contexts.
Stuart and Norman, I agree that ease of enforcement should be an issue, but surely there should be limits to how incoherent a concept we will use to make enforcement easy.
Re the patent example, I'd second Stuart Armstrong's point that the distinctions are based largely on ease of enforcement, and more particularly evidentiary concerns. The basic question in patent validity is whether the invention required the lure of a patent to bring it forth. If anyone made it public without bothering to apply for a patent, this is good evidence that a patent wasn't really required. Of course, it's not conclusive evidence as the disclosure might just have been inadvertent, but you have to draw the line somewhere.
Incidentally, while the US has a one-year grace period for disclosure by the inventor (you can get get a patent if the invention was disclosed by you in the year prior to the application date, but not if it was disclosed by someone else), European countries generally do not. I believe that the Cohen / Boyer patent, which is the basic patent on GM technology, was granted in the US but not in Europe because they had published an academic article disclosing the invention prior to applying for the European patent. Another interesting example is that the windsurfer was not patented in the UK because a 12 year-old boy made an used one for a summer at the family cottage in 1958: see Windsurfing International Inc v Tabur Marine (Great Britain) Ltd [1985] RPC 59(CA).
An invention might never have been disclosed and yet be obvious, in the sense of not requiring the lure of a patent for its development. The classic example is the automobile: it was obvious to mount an engine on a cart as soon as the engines became light enough. Such an invention is not patentable in principle as being obvious, so it is not really right to say that "you can keep patent rights even if lots of other people said they already thought of your obvious idea." Non-obviousness is indeed proven by testimony of experts in the field that the invention would have been obvious to a person with ordinary skill in the art. However, non-obviousness is considerably more difficult to establish than lack of novelty, as the courts are very aware that a truly worthy invention may appear obvious with the benefit of hindsight. And indeed, a Mr. Selden did get a US patent on the automobile, to his considerable profit.
Re insider trading, the Journal of Legal Studies Vol 9(4) (1980) is a special issue on the law and economics of privacy, and includes a couple of papers on insider trading.
Robin, great post.
"If insiders are required to preannounce trades, doesn't this make the info about preannouncement also insider information & allow for explotation?"
Of course the "preannouncement" information becomes insider information, but how do you suggest it could be exploited? It would be impossible for this new insider--the one with inside preannouncement information--to trade based on the new information unless he preannounces his intention.
Stuart: if you wanted such a fund to be successful, you could test how much insider information was dispersed by looking at the stock's performance after earnings surprises. In general, the way to measure how much insider information spreads is to see how the stock reacts to newly-public data. If an executive claims to be registering his trades with the fund, but the stock drifts upwards the month before a good earnings announcement, you could presume some amount of chicanery.
Fergus, great question. Husband of a lawyer? 'One financial expert'? Robin's reported source and linked confirmation gives me the sense this may be a protean urban myth.
"Legal definitions of insider trading have expanded beyond all common sense in recent years. One financial expert told me not long ago that I would be guilty of it if I saw a factory on fire from my plane as it landed in Des Moines, inquired about it, learned that the factory was owned by a company in which I had stock, and promptly telephoned my broker to sell before the news reached Wall Street."
Other alternative: what about funds, tracking exactly the purchases of insiders (and no restrictions on what insiders can buy). That way, wary investors could invest in these funds as a sort of "insurance against insider dealing", and academics could finally get an idea of what the information advantage on insiders really is.
Still not sure how to deal with proxy buying and passing info on to friends, though.
It seems "public" vs "private" in laws are based on mainly on ease of enforcement, rather than on coherent concepts.
The idea of "classifying traders into various levels of how informed they are" is a much better concept, but may suffer from enforcement problems. The major one concerns the children, friends and relatives of the "highly informed" traders. Traders may pass on information to them or use them as proxies. How to deal with this issue, without resurrecting some concept of "public" and "private" information?
On a side note, insider trading laws (just by their existence, not by the fact they do much) go a long way to improving the image of financial capitalism in general, and seem to encourage people to invest (especially those "overconfident investors" who contribute a lot to making the markets informative). Tinkering should be done with care, lest these advantages be lost.
BTW, did a search on Google, and haven't found any primary source for that factory fire story - it would be good to know more of the details, as it seems very pertinent to the issue. But it sounds like it might be an urban legend.
Fergus, I heard that one from the husband of a lawyer. Did I get it wrong? One source agrees with me.
Tuukka, yes info can often be "roughly" classified, but rough classifications do not support sharp legal distinctions well. I don't understand your concern about preannouncement.
There is information within organizations (firms, gov. bodies, etc.) that can be roughly classified as confidential or public. My understanding is that insider information is confidential information. That is, access to it is limited to some group by a contract or a law. Many ppl argue normatively that inverstors in the markets should have equal ground regarding the access to information. It doesn't seem good idea to trade with people with insider information, since on the average (& other things equal) you are losing.
When you are classifying traders in to various categories according to how informed they are, are't you implicitly making a categorization with respect to information as well?
If insiders are required to preannounce trades, doesn't this make the info about preannouncement also insider information & allow for explotation?
Maybe the solution could be "private law" or just firm-specific contracts which regulate insider trading. The solution might be something near present situation. To make the stock of company attractive for traders you might want to try exclude insiders from the market (also: some kind of insider regulation removes the incentive for bad management decisions by managers associated with gains in the stock market)
When and where did "insider trading laws punish ... a man who traded on the "private" info of seeing a factory fire when looking out an airplane window" ?