I’m a professor of economics who has published on politics, and in ten weeks the US will elect a president, and congress-folk. So a reasonable test of my supposed specialized knowledge is: do I have anything clearly useful to tell ordinary folks about how to vote in this election?
But the real situation is that typical voters struggle to reliably estimate *any* differing consequences of who they vote for. It is to such voters that I address my advice.
But it truly is bad advice! Voters ought to keep struggling until they can articulate a reliable difference. That's better than using your misleading measure.
All but a very few voters know anything significant about the people they vote for or why. They typically just adopt whatever political stance helps them feel they belong in their preferred social environment.
That's why you can't reason with them. Reason had nothing to do with their choices and they don't even know what reasoning really is, much less how to vet their sources of information. You might as well be speaking Klingon to these people when you try to reason with them.
That's what happens when you have an educational system full of teachers who can't teach critical thought because they don't have a clue how to think critically themselves. They just shove lots of rote learning down the poor kids' throats and believe that's what makes you smart...knowing lots of little facts that you just have to accept on authority. That is death to critical reasoning.
Right! This guy is a professor of economics? He doesn't get that it doesn't only matter how things were when a new president takes office, but also which way the economy was headed and how fast on day one? This is not to mention the delay between legislation and implementation, then the long delay between implementation of the policies and practical impact on a national scale. I guess he thinks aircraft carriers make U-turns in a couple of minutes, too. He must be a professor of economics in his children's home school!
Alas, we still don’t have presidential decision markets, that estimate important policy outcomes, such as GDP, unemployment, oil prices, etc., conditional on who becomes president. It would only take a few tens of thousands of dollars for someone to create these, at say Intrade.
In retrospect, weren't you overly optimistic about Intrade's entry barriers being inconsequential?
There are plenty of reasons beyond this not to have voted for Romney. Fortunately, a bare majority in enough states got that so he is not in the White House and never will be. Thank God!
This an amusing article with some deep truths embedded in it. The most important truth is that, as the author has stated, most people don't vote on the basis of anything rational regarding outcomes, but rather on what makes them feel they fit well and comfortably into their preferred social environment.
One really big truth the author misses, however, is that in this country we have developed a very strong class system based on wealth or lack of it that provides the moneyed elite with a very powerfully overbalanced influence on politics. These, on the whole, have a short-sighted wealth extracting agenda, which is to say they wield their influence to extract as much wealth from the rest of us as they can get government to go along with and reward their puppets in D.C. handsomely for taking their orders seriously.
This is economic suicide in the long run, but this agenda is either blind to that because of simple, short-sighted stupidity or an immoral attitude best expressed as "I'll be gone, you'll be gone" when the proverbial stuff hits the fan. They either are or can hire experts, however, to convince many that they are the only qualified saviors of the economy. After all, look how well they've done, never mind that a lot of it comes straight out of YOUR pocket. So they sucker many of their victims into voting for their political puppets to these voters' own and everyone else's great harm.
A second big truth this article misses is that economic policy takes just about four years to have any significant influence. So blaming Obama, for example, for the economic woes so many have suffered during his presidency is tantamount to blaming him for the sun's going down in the evening. If Romney is elected, he gets to inherit an economy that is about where it was when Obama took over, but headed in the opposite direction, even if much more slowly than it was going down when Obama became president. Taking credit for this would be tantamount to giving Romney credit for the sun's rising in the morning.
It is well known that the direction the economy is heading during an administration, up or down, is the most important predictor of an incumbent's chances of winning. Given the delay between policy and its effects roughly four years later, the obvious conclusion is that we mostly vote on the basis of irrelevant data. Both parties know this, of course, but also know that most people have no idea, so they exploit this ignorance to the fullest. Don't be a sucker. Don't keep voting against your own interests.
It's only strange if you make the assumption that Robin's goal in offering this advice is to help people make rational decisions about the election according to their own principles.
Wishes aren't horses. Yes it would be nice if voters knew enough to predict long term differing consequences of different candidates being elected. But the real situation is that typical voters struggle to reliably estimate *any* differing consequences of who they vote for. It is to such voters that I address my advice.
follow the link at the word "presidential" above to see how to deal with causation issues here.
It's a clever effort at a solution, but I wouldn't use it to tell me how to vote because I don't want to vote for the candidate who would be best conditional on the election being close. (And I don't think most people would want to vote that way.)
The argument that I should care particularly when the election is close goes that my vote "counts" the most then--which is to say, that's when the probability will be greatest that my vote will be decisive. But the probability of my vote's being decisive is so small, that few people actually base their vote on it.
If you can come up with a model that can simulate the entire economic universe and thus predict exactly what the economy is going to be like, congrats, you are now the richest person in human history. The 'actual' prediction market has fails at doing this every day. The 'high degree of uncertainty' is high enough that even if we could model and predict one man's decisions on countless economic institutions - which the field of economics + wall st is nowhere close to being capable of doing reliably - his effect is still marginal enough in the face of that uncertainty that it still doesn't become something that can be seen in the numbers. Maybe if a 5 year old boy who promised to make work illegal was one of the two candidates, then yes, his effect would overwhelm the extremely high market uncertainty about the future. The closest thing to a candidate who's crazy enough that the market could predict the American economy collapsing if he were elected would prob be Ron Paul. The stock market *would* be predicting that if it looked like it would happen - so we don't have to 'build a prediction market'. But candidates promising fairly marginal institutional changes yet who will be facing black swan challenges? Of course no one can 'predict their effect' on the thing markets already (conclusively) can't predict.
@google-1d23fb975fff3a0bbcecccb059cc9c63:disqus I don't understand your example. Let's say the economy is doing badly under Obama and the prediction market indicates it will continue to do badly if Romney is elected. But it will also indicate how the economy will do if Obama is reelected. How is this unfair to Romney?
The prediction market will be unfair to Romney when it predicts the economy will do better if Obama is re-elected but the better performance is predicted under Obama due to the conditions that caused Obama to win.
Whether the economy will do better if Obama is re-elected just isn't the same question as whether Obama will cause the economy to do better.
stephenrdiamond - "to use that prediction as a voting recommendation would confuse correlation and causation. For example, the prediction markets might correctly predict more unemployment under a Republican, but the reason might be that the Republican is more likely to win if the economy is doing badly under the Democrat"
I don't understand your example. Let's say the economy is doing badly under Obama and the prediction market indicates it will continue to do badly if Romney is elected. But it will also indicate how the economy will do if Obama is reelected. How is this unfair to Romney?
Why wouldn't a prediction market be able to communicate a high degree of uncertainty? Also, why isn't the president's effect on the economy a variable that can be isolated? I think I have a way to do it - simulate 2 universes, one where candidate A is elected and one where candidate B is elected. Then compare.
But the real situation is that typical voters struggle to reliably estimate *any* differing consequences of who they vote for. It is to such voters that I address my advice.
But it truly is bad advice! Voters ought to keep struggling until they can articulate a reliable difference. That's better than using your misleading measure.
All but a very few voters know anything significant about the people they vote for or why. They typically just adopt whatever political stance helps them feel they belong in their preferred social environment.
That's why you can't reason with them. Reason had nothing to do with their choices and they don't even know what reasoning really is, much less how to vet their sources of information. You might as well be speaking Klingon to these people when you try to reason with them.
That's what happens when you have an educational system full of teachers who can't teach critical thought because they don't have a clue how to think critically themselves. They just shove lots of rote learning down the poor kids' throats and believe that's what makes you smart...knowing lots of little facts that you just have to accept on authority. That is death to critical reasoning.
Right! This guy is a professor of economics? He doesn't get that it doesn't only matter how things were when a new president takes office, but also which way the economy was headed and how fast on day one? This is not to mention the delay between legislation and implementation, then the long delay between implementation of the policies and practical impact on a national scale. I guess he thinks aircraft carriers make U-turns in a couple of minutes, too. He must be a professor of economics in his children's home school!
Alas, we still don’t have presidential decision markets, that estimate important policy outcomes, such as GDP, unemployment, oil prices, etc., conditional on who becomes president. It would only take a few tens of thousands of dollars for someone to create these, at say Intrade.
In retrospect, weren't you overly optimistic about Intrade's entry barriers being inconsequential?
There are plenty of reasons beyond this not to have voted for Romney. Fortunately, a bare majority in enough states got that so he is not in the White House and never will be. Thank God!
This an amusing article with some deep truths embedded in it. The most important truth is that, as the author has stated, most people don't vote on the basis of anything rational regarding outcomes, but rather on what makes them feel they fit well and comfortably into their preferred social environment.
One really big truth the author misses, however, is that in this country we have developed a very strong class system based on wealth or lack of it that provides the moneyed elite with a very powerfully overbalanced influence on politics. These, on the whole, have a short-sighted wealth extracting agenda, which is to say they wield their influence to extract as much wealth from the rest of us as they can get government to go along with and reward their puppets in D.C. handsomely for taking their orders seriously.
This is economic suicide in the long run, but this agenda is either blind to that because of simple, short-sighted stupidity or an immoral attitude best expressed as "I'll be gone, you'll be gone" when the proverbial stuff hits the fan. They either are or can hire experts, however, to convince many that they are the only qualified saviors of the economy. After all, look how well they've done, never mind that a lot of it comes straight out of YOUR pocket. So they sucker many of their victims into voting for their political puppets to these voters' own and everyone else's great harm.
A second big truth this article misses is that economic policy takes just about four years to have any significant influence. So blaming Obama, for example, for the economic woes so many have suffered during his presidency is tantamount to blaming him for the sun's going down in the evening. If Romney is elected, he gets to inherit an economy that is about where it was when Obama took over, but headed in the opposite direction, even if much more slowly than it was going down when Obama became president. Taking credit for this would be tantamount to giving Romney credit for the sun's rising in the morning.
It is well known that the direction the economy is heading during an administration, up or down, is the most important predictor of an incumbent's chances of winning. Given the delay between policy and its effects roughly four years later, the obvious conclusion is that we mostly vote on the basis of irrelevant data. Both parties know this, of course, but also know that most people have no idea, so they exploit this ignorance to the fullest. Don't be a sucker. Don't keep voting against your own interests.
It's only strange if you make the assumption that Robin's goal in offering this advice is to help people make rational decisions about the election according to their own principles.
My golden retriever tells me how to vote. This November I'm writing in, "bacon".
Wishes aren't horses. Yes it would be nice if voters knew enough to predict long term differing consequences of different candidates being elected. But the real situation is that typical voters struggle to reliably estimate *any* differing consequences of who they vote for. It is to such voters that I address my advice.
follow the link at the word "presidential" above to see how to deal with causation issues here.
It's a clever effort at a solution, but I wouldn't use it to tell me how to vote because I don't want to vote for the candidate who would be best conditional on the election being close. (And I don't think most people would want to vote that way.)
The argument that I should care particularly when the election is close goes that my vote "counts" the most then--which is to say, that's when the probability will be greatest that my vote will be decisive. But the probability of my vote's being decisive is so small, that few people actually base their vote on it.
If you can come up with a model that can simulate the entire economic universe and thus predict exactly what the economy is going to be like, congrats, you are now the richest person in human history. The 'actual' prediction market has fails at doing this every day. The 'high degree of uncertainty' is high enough that even if we could model and predict one man's decisions on countless economic institutions - which the field of economics + wall st is nowhere close to being capable of doing reliably - his effect is still marginal enough in the face of that uncertainty that it still doesn't become something that can be seen in the numbers. Maybe if a 5 year old boy who promised to make work illegal was one of the two candidates, then yes, his effect would overwhelm the extremely high market uncertainty about the future. The closest thing to a candidate who's crazy enough that the market could predict the American economy collapsing if he were elected would prob be Ron Paul. The stock market *would* be predicting that if it looked like it would happen - so we don't have to 'build a prediction market'. But candidates promising fairly marginal institutional changes yet who will be facing black swan challenges? Of course no one can 'predict their effect' on the thing markets already (conclusively) can't predict.
In California, you receive a stub from the ballot, and usually, a sticker which says "I voted".
Isn't there an option not to express any preference?
You can demonstrate your alligiences by getting into internet discussions about politics where your friends or coworkers will see them.
But that doesn't necessarily reflect how and whether you vote.
Guys, follow the link at the word "presidential" above to see how to deal with causation issues here.
@google-1d23fb975fff3a0bbcecccb059cc9c63:disqus I don't understand your example. Let's say the economy is doing badly under Obama and the prediction market indicates it will continue to do badly if Romney is elected. But it will also indicate how the economy will do if Obama is reelected. How is this unfair to Romney?
The prediction market will be unfair to Romney when it predicts the economy will do better if Obama is re-elected but the better performance is predicted under Obama due to the conditions that caused Obama to win.
Whether the economy will do better if Obama is re-elected just isn't the same question as whether Obama will cause the economy to do better.
stephenrdiamond - "to use that prediction as a voting recommendation would confuse correlation and causation. For example, the prediction markets might correctly predict more unemployment under a Republican, but the reason might be that the Republican is more likely to win if the economy is doing badly under the Democrat"
I don't understand your example. Let's say the economy is doing badly under Obama and the prediction market indicates it will continue to do badly if Romney is elected. But it will also indicate how the economy will do if Obama is reelected. How is this unfair to Romney?
Why wouldn't a prediction market be able to communicate a high degree of uncertainty? Also, why isn't the president's effect on the economy a variable that can be isolated? I think I have a way to do it - simulate 2 universes, one where candidate A is elected and one where candidate B is elected. Then compare.