The Top 1% is Stalling the Economy

[TimG]

What I also said is that the GINI predicts those social and health metrics. If you know what a country's GINI index is, you can make predictions about where their health and social metrics are.

Who cares? Predictors are only useful if the actual data is not available. In this case the actual data is available which makes the GINI useless. This has been my argument all along.

[cybercoma]

It isn't about giving politicians money to spend; it's about giving consumers money to spend, which signals the producers to create jobs.

[TimG]

It isn't about giving politicians money to spend; it's about giving consumers money to spend, which signals the producers to create jobs.

And if consumers spend that money on imported goods then there are no jobs created in Canada. Middle class consumers are just as fickle as big business when it comes to getting the best price no matter what the cost to the country.

But more importantly, you cannot get the money to consumers without passing it through big government first which would redirect a lot of the money to buying off unions and various exercises in crony capitalism. I have absolutely no faith that government could do anything useful with the money even if it took it.

[cybercoma]

Who cares? Predictors are only useful if the actual data is not available. In this case the actual data is available which makes the GINI useless. This has been my argument all along.

It's a predictor because they are connected in some way, Tim. Something is causing them to move together. The obvious connection shows that people with more money are healthier. Increasing how much the poorest have in the more unequal economies raises their health, thereby increasing the national averages. You see one word, "predictor," and it's like you didn't understand a thing I was saying otherwise.

[TimG]

It's a predictor because they are connected in some way, Tim. Something is causing them to move together.

Now you are assuming causality when it could be nothing but a random correlation. Like I said, dig through enough data you will find stuff that correlates by chance. The collection of metrics which correlate to the GINI appears to be pretty random.

A properly designed analysis would choose the collection of metrics before comparing to the GINI and once the comparison was done the list of metrics could not change. But that is not how researchers work - they hunt around looking for the most compelling data to make the point they want to make. Such an approach makes for more marketable headlines but useless science.

[Michael Hardner]

Too much equality isnt a problem as long as your means to get there isnt overly coercive.

Economists disagree with you on that one; material gain is an incentive to excel and to produce more. I think that that is self-evident. Post WW2 the GINI coefficient in America was roughly where it is in Canada today - between 35 and 40. Around 1980, the American GINI started going up.

http://www2.econ.iastate.edu/classes/econ521/orazem/inequality%20information.pdf

But there are economic arguments against the GINI being in the mid 20s as with Sweden, Norway, Denmark and Finland.

[cybercoma]

Now you are assuming causality when it could be nothing but a random correlation.

First of all, I still haven't said anything about GINI causing higher health metrics. I said they move together through the relationship between personal income and health at the individual level. GINI tells us nothing about individuals and neither do national health metrics. They only hint at those things. At the individual level, people with higher personal incomes tend to be healthier. That is a causal relationship that has been examined through longitudinal data analysis. People that are in poverty, who cannot provide the necessities of life to themselves, but then get back on their feet and have a better income are healthier than those who don't. That's not even a controversial relationship. This relates to GINI in the way that I mentioned earlier. If we as a society decide that we need to keep people from poverty and raise the lowest end of the spectrum, they will be healthier and closer in health to those at the higher end. This raises the nation's average and also closes the gap in term of GINI. That's one of many possible reasons that GINI has a negative relationship with health metrics (as inequality exists health stats go down).

Secondly, correlation is not random. With all due respect, it's pretty near impossible even discussing this with you if you know so little about how these studies are conducted that you would call correlation "random".

[cybercoma]

Economists disagree with you on that one; material gain is an incentive to excel and to produce more. I think that that is self-evident. Post WW2 the GINI coefficient in America was roughly where it is in Canada today - between 35 and 40. Around 1980, the American GINI started going up.

http://www2.econ.iastate.edu/classes/econ521/orazem/inequality%20information.pdf

But there are economic arguments against the GINI being in the mid 20s as with Sweden, Norway, Denmark and Finland.

Absolute equality is a problem for the reasons you say, as is a very high amount of inequality. It's about striking a balance.

There's absolutely no reason, however, that a CEO should earn 2000+ times that of the average employee in his/her company.

Edited January 7, 2014 by cybercoma

[TimG]

There's absolutely no reason, however, that a CEO should earn 2000+ times that of the average employee in his/her company.

Except there is not a damn thing the government can do about that because CEOs will simply structure their pay packets to get around any rules. CEO pay packets are a function of cultural values and as much as I would like them to change they are not going to simply because the government mandates it. Edited January 7, 2014 by TimG

[TimG]

Secondly, correlation is not random. With all due respect, it's pretty near impossible even discussing this with you if you know so little about how these studies are conducted that you would call correlation "random".

The definition of correlation means that it is 95% likely that the relationship is not random. That ALSO means that it there is a 5% probability that it is random. This, of course, assumes the researchers did not introduce selection bias where they sift though a large number of variables looking for the ones that correlate. If the researchers did do this then the possibility of a random correlation is close to 100%.

You need to learn what correlation analysis and, more importantly, what its limitations are.

Edited January 7, 2014 by TimG

[cybercoma]

That's not what correlation is at all. What you're about is an alpha level, as chosen by a researcher before running a regression analysis, to determine whether or not the null hypothesis will be accepted or rejected. That's not what correlation means and it sure as hell doesn't mean that correlation is random.

[cybercoma]

In any case, that has nothing to do with your point that GDP is a better predictor of these things than GINI. Now you're just trying to argue that the entire notion of statistical analysis is bunk because they use significance levels. Needless to say, you're being ridiculous.

[TimG]

That's not what correlation is at all. What you're about is an alpha level, as chosen by a researcher before running a regression analysis, to determine whether or not the null hypothesis will be accepted or rejected. That's not what correlation means and it sure as hell doesn't mean that correlation is random.

Yes it does. The p-value is the normal way to measure correlation and it means is there is a chance that the correlation was random. In some cases researchers can choose higher or lower p-values but 95% is the norm. But these probabilities mean nothing if the variables were selected by sifting through a much larger dataset. If data mining is done the relationship is likely random.

Lets put it this way. Lets say you have a 100 variables that measure social outcomes. If you test each one of them for correlation with some other variable then chances are 5 of the variables will correlate with a p-value > 95%. If you then go write a paper about the importance of those 5 variables you can make great headlines with science that is fundamentally bogus.

Edited January 8, 2014 by TimG

[cybercoma]

Yes it does. The p-value is the normal way to measure correlation and it means is there is a chance that the correlation was random. In some cases researchers can choose higher or lower p-values but 95% is the norm. But these probabilities mean nothing if the variables were selected by sifting through a much larger dataset. If data mining is done the relationship is likely random.

Lets put it this way. Lets say you have a 100 variables that measure social outcomes. If you test each one of them for correlation with some other variable then chances are 5 of the variables will correlate with a p-value > 95%. If you then go write a paper about the importance of those 5 variables you can make great headlines with science that is fundamentally bogus.

Tim, I know everything there is to know about p-values.

You keep saying that the significance levels are at .05; however, that's simply the level at which it's accepted that the particular statistic is not random. More importantly, that level is chosen by the researcher and is not necessarily 5%. Even when the significance level is chose as 5%, that does not mean that the probability in the study is at that. In fact, it's often much lower than that. In most social science studies they will tell you the p-value of a finding. Most tables will list the p-values and when they don't they'll use a number of asterisks to indicate whether it's significant at <.05, <.01, or <.001, regardless of what the particular researcher chooses as their alpha level.

Since you're so hung up on this idea of p-values and seem to think you have a valid argument against this GINI discussion. Tell me what the p-values were in whatever it is that you think you're trying to discredit.

This is nothing more than a diversion into methodology that you only seem to have a loose grasp on and has nothing to do with what we're actually discussing here. What we're actually discussing is clear. Put GDP on the y-axis or GINI on the y-axis of a chart. Then put your social measures on the X-axis. In wealth nations, when you plot the points (we're not talking regressions here or some other statistical analysis) you can see at a glance that it is a scattered mess with GDP as a predictor. When you use GINI the points line up and show a clear relationship. This has nothing to do with significance levels or p-values or regression modelling or ANOVA or whatever else you want to pull out of your 30 second wiki search for stats terms.

Further still, what makes your point entirely meaningless is that we're not using samples. Significance levels have to do with taking a sample and making generalizations about a population. What is the chance that our sample represents the entire population? That's what p-value determines. We're not taking a sample of nations. We're using the actual values for every nation in the OECD here. When you do that it's clear that there is less of a relationship between GDP and the outcomes we're discussing than there is with GINI.

[dre]

Economists disagree with you on that one; material gain is an incentive to excel and to produce more. I think that that is self-evident. Post WW2 the GINI coefficient in America was roughly where it is in Canada today - between 35 and 40. Around 1980, the American GINI started going up.

http://www2.econ.iastate.edu/classes/econ521/orazem/inequality%20information.pdf

But there are economic arguments against the GINI being in the mid 20s as with Sweden, Norway, Denmark and Finland.

The incentive to produce more does not mean require the economic framework to allow concentration of wealth. No matter how much you tax property you will still get more property by working hard than if you dont. No matter how much of your fortune the government puts back in the pool after you die, you will still have had more stuff if you work hard.

BTW, those countries have some of the best standards of life in the history of the human race. And they also still have the profit motive, and wealthy people.

The question is whether or not we want to live in an aristocracy where a small percentage of people have ALL the wealth. That not going to help the economy in any way at all, and we have a situation now where its not productive behavior that creates wealth its simply the act of having wealth. An auto mechanic pays more taxes on his dollar than wealthy billionaire investor.

You obviously dont want to force an absolutely even distribution of wealth, but beyond that all the most successful economies in the world have been the most egalitarian ones, and the ones that concentrate wealth the least.

[dre]

A bigger fallacy is the belief that our social problems could be solved if we only gave our wise and benevolent politicians more money to spend. Bigger government is never the solution - it is the problem.

If you think the size of government has anything to do with taxes you are kidding yourself. Its the exact opposite. When top marginal tax rates were at 70+% and wealth was deconcentrating the US government was half the size it is today as a percentage of GDP. Government has grown as wealth CONCENTRATED, because so many people are left fighting for table scraps while a tiny percentage hoards all the wealth. So they vote for government programs to help them.

If you had a more egalitarian economy you could shut most of the government down.

So yeah... you have it exactly backwards.

[Bonam]

If you had a more egalitarian economy you could shut most of the government down.

But the way to make the economy more egalitarian is not necessarily to give politicians more of our tax money to squander.

[bush_cheney2004]

Everybody who wants a more "egalitarian economy" is free to donate or invest personal wealth in the cause. Instead, they want a tax on somebody else's wealth.

[Michael Hardner]

The incentive to produce more does not mean require the economic framework to allow concentration of wealth. No matter how much you tax property you will still get more property by working hard than if you dont.

No, that's not true. Going back to your statement 'too much equality isn't a problem if...'. There is such a thing as too much equality, if you don't reward more difficult work, harder work with higher pay.

See Cyber's comment above also.

You obviously dont want to force an absolutely even distribution of wealth, but beyond that all the most successful economies in the world have been the most egalitarian ones, and the ones that concentrate wealth the least.

"Most egalitarian" does mean "even distribution of wealth".

Also, since we're talking about a coefficient the terms you've used "allow concentration of wealth" "some people have all the wealth" don't make sense. I suppose wealth is concentrated, somewhat, if you don't have perfectly even distribution. And there isn't a system in the world where some people have all the wealth.

Also keep in mind that I'm not advocating for any particular GINI target, I'm just reporting the conventional wisdom.

[Michael Hardner]

Everybody who wants a more "egalitarian economy" is free to donate or invest personal wealth in the cause. Instead, they want a tax on somebody else's wealth.

Or they can vote in a regime to change the rules of the game. The easiest thing in the world, really.

[bush_cheney2004]

Or they can vote in a regime to change the rules of the game. The easiest thing in the world, really.

That's the same thing as confiscating wealth. Nice try.....

[Michael Hardner]

That's the same thing as confiscating wealth. Nice try.....

It sure is. See America's global dominance from mid-20th century for an examination of the results.

[TimG]

Its the exact opposite. When top marginal tax rates were at 70+% and wealth was deconcentrating the US government was half the size it is today as a percentage of GDP.

There was also no government funding for healthcare or retirements. No environmental regulation.

If you think those programs are going to be rolled back simply because tax rates are raised then you are dreaming. But it is these kinds of entrenched interests that ensure that any new tax revenue that the government gets today will not go to the 'middle class' but will instead be diverted to line the pockets of the relatively wealthy people and corporations with political connections.

Edited January 8, 2014 by TimG

[TimG]

You keep saying that the significance levels are at .05; however, that's simply the level at which it's accepted that the particular statistic is not random.

No. It is the level at which the researcher *assumes* that it is not random. Even at that level there is still a 5% chance that it is actually random even if the researcher assumes otherwise.

But these 5% or 10% or 1% probabilities only apply if the researcher does not introduce selection bias by mining data for variables that happen to correlate because the researcher can easily discard the 95% of the data that does not correlate and only keep the 5% that does correlate. This is why I say the chances of random correlation goes up significantly when data mining goes on. This chance can be reduced if there is some independent evidence that the correlation should exist but in this case there is no independent evidence - all of the evidence is the correlation.

Further still, what makes your point entirely meaningless is that we're not using samples.

Wrong. At a basic level all data is sampled (i.e. a discrete series of measurements over time). If you want to a physical measure like tree rings and use it to predict temperatures then you use p-levels to show that the relationship between your available tree ring data and temperature is not random. However, researchers in this field also introduce selection bias by discarding samples that don't correlate which renders their claims meaningless. But that does not stop them from making strong claims.

It is very easy to fool yourself with stats. You appear to like being fooled when it supports your pre-determined ideas.

Edited January 8, 2014 by TimG

[bush_cheney2004]

It sure is. See America's global dominance from mid-20th century for an examination of the results.

Seriously ? Now this is mostly an American issue ? Did you expect or prefer "global dominance" by Canada ?