Empirically testable right-wing beliefs

[Darth Wong]

Axis Kast has stated many times before that many right-wing beliefs are subject to empirical testing and that this is probably the most effective way to debate them. I would tend to concur. Therefore, it may be useful to create a list of empirically testable right-wing beliefs.

Economics

"Welfare creates disincentive to work": this belief is provocative and attractive on an intuitive level. However, if it were true, we would expect to see that the majority of welfare recipients make no effort to gain employment. We should also expect to see that the majority of welfare recipients stay on welfare for as many years as possible, rather than getting off welfare in two years or less.

"Reducing tax rates for the wealthy will increase economic growth for all, which will benefit the other classes": this right-wing belief generates numerous testable predictions, one of which is that there should also be an inverse correlation between Wall Street wealth and working class unemployment or underemployment.

"Government-run health care would create a bloated, massively inefficient socialist bureaucracy": if this belief were true, then countries with socialized health-care should have extraordinarily expensive health care systems, extremely low quality health care, or some combination of both. It would be impossible, given this belief, for a country with a socialized health care system to produce equal quality of care at equal or lower cost per capita.

"Social programs could be replaced more equitably and efficiently by private charity": if this were true, then we should expect to see that private charity steps in to fill the gap when government programs are absent. Therefore, in the United States which presently lacks universal health care, we should expect to see that private charity completely fills that void and provide health care to the working poor who currently make up the bulk of the uninsured. We should also expect to see that private charitable donations increase during times of economic hardship, because that is when the need is greatest, and that is when a government social program would normally spend more.

"The greed of trade unions will eventually make our industries uncompetitive and sap the strength of our economy": if this is true (and I personally think it actually is), then we should expect to see the most heavily unionized industries experience serious problems with efficiency and competitiveness, not just relative to foreign countries but also relative to other less-unionized industries in the same country.

Crime

"Tougher criminal sentencing will deter crime": this belief leads to the testable prediction that regions which implement tougher sentencing rules will experience a drop in crime rates and recidivism rates.

"The death penalty is an effective deterrent to violent crime": similar to the previous example: we should see violent crime rates decrease once a death penalty is instituted, or increase when the death penalty is outlawed.

Social Policy

"Abstinence-only sex education reduces teen sexual activity, thus reducing teen pregnancy": if this is true, we should expect to see reduced rates of teen sexual activity in areas which use abstinence-only sex education. One would also expect to see reduced rates of teen pregnancy.

"Legalizing gay marriage would set us on a slippery slope toward the destruction of all values and morals": if this is true, we should expect to see crime rates increase in any state or nation which has legalized gay marriage. Retroactively, it also predicts that the historical abolition of anti-sodomy laws should have generated a sharp increase in crime.

"Legalized abortion on demand encourages women to use abortion as a form of contraception": this right-wing belief generates the testable prediction that women will decrease their use of other forms of contraception once abortion becomes legally available.

"When governments grow in size and power, tyranny inevitably follows": if this is true, it should be possible to look at all of the world's nations (not just cherry-picked examples from history) and detect a reliable correlation between size of government and loss of social freedoms such as freedom of speech and freedom of association.

"When women wear revealing clothes, they make themselves into objects, and men are more likely to sexually assault them because they no longer see them as people": if this is true, then we should expect to see that sexual assault rates from different nations around the world are correlated with the popularity of revealing clothes in each nation's culture.

"The strength of marriage derives from the strength of the couple's faith in God": if this is true, then we should find an inverse correlation between religious faith and divorce rates (or spousal abuse rates, since one could hardly argue that a marriage filled with violent assaults is a "strong" marriage).

Foreign Affairs

"Diplomatic contact with our enemies will legitimize and embolden them": this right-wing belief generates the testable prediction that a complete cessation of diplomatic contact will have a relatively intimidating effect on enemies, making them more likely to grant concessions.



Any other examples? Any good resources on data to test these predictions?

[darthdavid]

Legalizing Drugs will lead to a dramatic increase in drug use and cause significantly more harm to the country than the war on drugs currently causes.

[Samuel]

"Legalized abortion on demand encourages women to use abortion as a form of contraception": this right-wing belief generates the testable prediction that women will decrease their use of other forms of contraception once abortion becomes legally available.

Freakanomics mentions that this is true- there was an increase in conceptions after Roe v Wade, but a drop in the birth rate. Of course I don't see how women using abortion on demand is a bad thing. But then this would be asking the wrong question...

Punishing people for getting pregnant reduces the rate at which people have sex IS the right one.

[Formless]

"Diplomatic contact with our enemies will legitimize and embolden them": this right-wing belief generates the testable prediction that a complete cessation of diplomatic contact will have a relatively intimidating effect on enemies, making them more likely to grant concessions.

Wait a second... what kind of complete retard thought this one up? It doesn't even fulfill a basic logic test, let alone empirical testing. It proposes that we can both stop communicating with our enemies AND expect them to be more willing to grant us concessions... which they can't grant because we aren't communicating with them. What the fuck? The only way that makes sense is if by "legitimize and embolden" they mean "we are ACTIVELY TRYING to start a war, and consider our enemies somehow less than human." Its completely devoid of everything rational or moral.

[Darth Wong]

"Diplomatic contact with our enemies will legitimize and embolden them": this right-wing belief generates the testable prediction that a complete cessation of diplomatic contact will have a relatively intimidating effect on enemies, making them more likely to grant concessions.

Wait a second... what kind of complete retard thought this one up?

FOXNews. The "legitimize and embolden our enemies" line is one of their favourite attacks against Obama.

It doesn't even fulfill a basic logic test, let alone empirical testing. It proposes that we can both stop communicating with our enemies AND expect them to be more willing to grant us concessions... which they can't grant because we aren't communicating with them. What the fuck? The only way that makes sense is if by "legitimize and embolden" they mean "we are ACTIVELY TRYING to start a war, and consider our enemies somehow less than human." Its completely devoid of everything rational or moral.

I presume they expect other nations to go to the UN and beg forgiveness because they are refusing to speak to them, in much the same way that a teenaged girl in high school might force a wayward boyfriend to apologize through a proxy while she gives him the silent treatment. Or perhaps they expect other nations to stand outside their windows while holding a ghetto blaster above their heads which is playing "In Your Eyes" by Peter Gabriel.

In any case, as imbecilic or immature as it may be, it still generates a potentially testable prediction.

[Formless]

FOXNews. The "legitimize and embolden our enemies" line is one of their favourite attacks against Obama.

I dunno. The exact line probably comes from Faux, but I think the basic mindset might actually be older than that. Anyone remember "we don't negotiate with terrorists"?

In any case, as imbecilic or immature as it may be, it still generates a potentially testable prediction.

No, because there is no mechanism for which the result they are proposing can possibly occur. Any test you try will only come back "this is impossible on the face of it."

[Darth Wong]

FOXNews. The "legitimize and embolden our enemies" line is one of their favourite attacks against Obama.

I dunno. The exact line probably comes from Faux, but I think the basic mindset might actually be older than that. Anyone remember "we don't negotiate with terrorists"?

In any case, as imbecilic or immature as it may be, it still generates a potentially testable prediction.

No, because there is no mechanism for which the result they are proposing can possibly occur. Any test you try will only come back "this is impossible on the face of it."

Not to defend such a childish belief, but that's not true. The "enemy" in question could unilaterally do whatever the United States has been demanding that they do, without having to negotiate with the US first. That is a concession by action, rather than words.

If I hold a gun to your head, tell you that you're not allowed to speak, and instruct you to hand over your money, you could hand over your money without actually being able to verbally negotiate with me.

[Formless]

True. I hadn't thought of that.

[Samuel]

FOXNews. The "legitimize and embolden our enemies" line is one of their favourite attacks against Obama.

I dunno. The exact line probably comes from Faux, but I think the basic mindset might actually be older than that. Anyone remember "we don't negotiate with terrorists"?

Actually the point of "we don't negotiate with terrorists" is that if you don't give into their demands you make terrorism less rewarding and people stop doing it. As you can imagine this only really works for situations like terrorists who make demands while holding hostages.

[Feil]

"Welfare creates disincentive to work": this belief is provocative and attractive on an intuitive level. However, if it were true, we would expect to see that the majority of welfare recipients make no effort to gain employment. We should also expect to see that the majority of welfare recipients stay on welfare for as many years as possible, rather than getting off welfare in two years or less.

Would we? Why draw the line at "majority"? Why not eighty percent? Why not thirty percent? If we could compare welfare recipients to a viable control group that received no welfare, maybe we could draw meaningful conclusions. Without it... things become much more difficult. For one, the claim is in a broad sense true, since it leaves out the magic words "statistically significant". The question that needs asking is how much disincentive to work does Welfare create, and to what extent, if any does that disincentive delay the recipient's acquisition of employment? It may be possible to measure this, but the method you postulate wouldn't produce conclusive results.

[Darth Wong]

A fair point. Perhaps a more thorough quoting of the original right-wing belief is in order: For example, "The Right has long held, and the Left is coming reluctantly to agree, that welfare creates a culture of dependency, sapping the initiative of its recipients. In the slums right now a generation of illegitimate children raised fatherless on Aid to Families with Dependent Children is being encouraged by welfare to produce the next generation." (from http://www.thefreemanonline.org/columns ... hould-end/ )

It has always been stated in such a manner as to strongly suggest that the "culture of dependency" effect can be generalized for the entire welfare population, which in turn implies that they refer to a majority, not just a barely measurable uptick in percentages.

[The Grim Squeaker]

Economics:
Making it harder for employers to fire employees, as well as increasing the minimum wage increases unemployment - it becomes more expensive and risky for employers to hire new workers if it means they can't fire them without suffering substantial losses.

(these are two seperate arguments leading to the same conclusion of increased unemployment. Evidence bought up for them is the higher levels of unemployment in Europe compared to the states, with a very tight causality between labor laws and employment rates in equilevent cases. Africa is bought up often as an argument on minimum wages, namely that if they were even lower than the companies would hire even more people, and the wages are still much, much higher than local standards and the work better).

[Dark Hellion]

Crime

"Tougher criminal sentencing will deter crime": this belief leads to the testable prediction that regions which implement tougher sentencing rules will experience a drop in crime rates and recidivism rates.

"The death penalty is an effective deterrent to violent crime": similar to the previous example: we should see violent crime rates decrease once a death penalty is instituted, or increase when the death penalty is outlawed.

These would not be a true test of deterrence as they assume an equal criminality across all social groups. This is clearly not the case, crime seeming to be more a function of socio-economic factors and cultural factors (gangsta culture or corporate culture for example). A more proper empirical test would require two groups with similar social, economic and cultural backgrounds with one featuring the deterrent to be tested (and only that deterrent).

Testing for deterrence at all is very hard as well because it assumes that criminals at the time of their crime are rational actors, where as I am sure that it can be demonstrated via a quick perusal of statistics that a significant portion are under some intoxicant that would prevent them from fully considering the consequences of their actions and thus negating some of the deterrent effect that would be present in a normal individual. Which must also be taken into account when deciding whether a deterrent effect is actually functioning properly or not.

Generally, this is why such debates go round and round in circles. Humans are clearly not perfectly rational actors and thus even the strongest deterrent will not necessarily keep them from committing an action, while at the same time they are rational enough actors that we should seem to be able to assume that fear of punishment will prevent actions that will result in punishment. So we have one side which says "Well people obviously won't commit crimes if they fear being punished" and another which cites statistics such as Texas crime statistics even though really both of these are sides of the same coin as deterrence alone is clearly not the only factor in crime prevention but rationally seems like it should be one of the contributing factors.

[Ariphaos]

"Government-run health care would create a bloated, massively inefficient socialist bureaucracy": if this belief were true, then countries with socialized health-care should have extraordinarily expensive health care systems, extremely low quality health care, or some combination of both. It would be impossible, given this belief, for a country with a socialized health care system to produce equal quality of care at equal or lower cost per capita.

I've dealt with arguments against this where people claim that the United States is more diverse than socialized Europe or Japan - so it's always good to point out Australia's system specifically, as serving a diverse, obese populace.

[Surlethe]

Some thoughts on economics right-wing claims.

"Welfare creates disincentive to work": this belief is provocative and attractive on an intuitive level. However, if it were true, we would expect to see that the majority of welfare recipients make no effort to gain employment. We should also expect to see that the majority of welfare recipients stay on welfare for as many years as possible, rather than getting off welfare in two years or less.

Feil has a good point about this, but I'd add that we could actually explicitly test this by considering panel data of different countries' unemployment rates as compared to their welfare benefits. A cursory google gives this; a graph of unemployment rates (one should also look at labor force participation rates, but I've taken too much time this morning as is):

There's almost certainly a large body of economic literature on this.

"Reducing tax rates for the wealthy will increase economic growth for all, which will benefit the other classes": this right-wing belief generates numerous testable predictions, one of which is that there should also be an inverse correlation between Wall Street wealth and working class unemployment or underemployment.

We might test this with historical data from the Gilded Age, especially before 1916. I suspect that there is a "natural" distribution of income, reached under the usual free-market conditions - log-normal or Poisson, perhaps - and so this would be true, but would of course imply always-higher growth rates for the tail of the spectrum. In any case, I'm sure there's a body of research on modeling income distributions and, hopefully, testing them - the paper you cited the other day on whether CEOs are overpaid cited some as references, IIRC.

"Social programs could be replaced more equitably and efficiently by private charity": if this were true, then we should expect to see that private charity steps in to fill the gap when government programs are absent. Therefore, in the United States which presently lacks universal health care, we should expect to see that private charity completely fills that void and provide health care to the working poor who currently make up the bulk of the uninsured. We should also expect to see that private charitable donations increase during times of economic hardship, because that is when the need is greatest, and that is when a government social program would normally spend more.

Another way to test this is retroactively, in the Gilded Age. I know some profs who might be able to get me some data for charitable giving; according to them, it plummeted in the mid-1930s when the modern US welfare system started to take place. Of course, that doesn't answer how charitable spending during the Gilded Age performs relative to a hypothetical welfare state's spending; one would have to take current policies and extrapolate them back.

"The greed of trade unions will eventually make our industries uncompetitive and sap the strength of our economy": if this is true (and I personally think it actually is), then we should expect to see the most heavily unionized industries experience serious problems with efficiency and competitiveness, not just relative to foreign countries but also relative to other less-unionized industries in the same country.

One would have to correct this for monopoly power: the fewer competitors in a market, the more inefficiency and uncompetitiveness a particular firm can get away with. For example, how much of GM's bloated structure was due to unions and how much to its lack of competition for fifty-some years?

Another one:
"Higher taxes create a disincentive to work, reducing growth rates": We can test this by looking at data from different countries with variable tax rates (measured as % of GDP) and see if there is the expected correlation in growth rates. We could also see if periods of higher tax rates in the past have led to decreased growth, and periods of lower taxes have led to increased growth (looking of course at all periods, not just one or two - no cherry-picking WWII, e.g.).

[Surlethe]

Damn, missed the timer. Here's the graph:



(N.B.: The BLS has corrected the data so it is all measuring the same (US) unemployment rate. The comparisons are valid.)

[Surlethe]

I recall from Conscience of a Liberal that Krugman dropped some information in about rise in median wages during what he calls the "Long Gilded Age" (1870-ish to 1930).

Anyway, here is a list of papers related to income distribution, cribbed from http://www.mdpi.com/1099-4300/11/4/766/pdf (the paper on CEO pay posted recently), if anybody has time to follow up on them:

• 6. Champernowne, D.G.; Cowell, F.A. Economic Inequality and Income Distribution; Cambridge University Press: Cambridge, UK, 1999.
  7. Pareto, V. Cours d’economie Politique; F. Rouge: Lausanne, Switzerland, 1897.
  8. Gini, C. Measurement of inequality and incomes. Econ. J. 1921, 31, 124-126.
  21. Souma, W. Universal structure of the personal income distribution. Fractals 2001, 9, 463-470.
  22. Bouchaud, J.P.; Mezard, M. Wealth condensation in a simple model of economy. Physica 2000, A282, 536-545.
  23. Levy, M.; Solomon, S. Power laws are logarithmic Boltzmann laws. Int. J. Mod. Phys. C 1996, 7, 595-601.
  24. Chatterjee, A.; Sinha, S.; Chakrabarti, B.K. Economic inequality: Is it natural? Curr. Sci. 2007, 92, 1383-1389.
  25. Richmond, P.; Hutzler, S.; Coelho, R.; Repetowicz, P. A Review of Empirical Studies and Models of Income Distributions in Society. In Econophysics and Sociophysics: Trends and Perspectives; Chakrabarti, B.K., Chakraborti, A., Chatterjee, A., Eds.; Wiley-VCH: Berlin, Germany, 2006. Entropy 2009, 11 781
  26. Gallegati, M.; Keen, S.; Lux, T.; Ormerod, P. Worrying trends in econophysics. Physica A 2006, 370, 1-6.
  27. Willis, G.; Mimkes, J. Evidence for the independence of waged and unwaged income, evidence for Boltzmann distributions in waged income, and the outlines of a Coherent theory of income distribution; arXiv:cond-mat/0406694, 2004.
  28. Sinha, S. Stochastic maps, wealth distribution in random asset exchange models and the marginal utility of relative wealth. Phys. Scr. T 2003, 106, 59-64.
  29. Chatterjee, A.; Chakrabarti, B.K. Kinetic market models with single commodity having price fluctuations. Eur. Phys. J. B 2006, 54, 399-404.
  30. Hayes, B. Follow the money. Am. Sci. 2002, 90, 400-405.
  31. Hogan, J. Why it is hard to share the wealth. New Sci. 2005, 2490, 6-8.
  32. Ball, P. Econophysics: Culture crash. Nature 2006, 441, 686.
  33. Econophysicists matter. Nature 2006, 441, 667.

I should also note that my comment immediately after the graph above is unclear: it should read, "N.B.: The BLS has corrected the unemployment statistics so they use the same definitions as the baseline US rate. The comparisons are valid."

[Gilthan]

"Higher taxes create a disincentive to work, reducing growth rates": We can test this by looking at data from different countries with variable tax rates (measured as % of GDP) and see if there is the expected correlation in growth rates. We could also see if periods of higher tax rates in the past have led to decreased growth, and periods of lower taxes have led to increased growth (looking of course at all periods, not just one or two - no cherry-picking WWII, e.g.).

I don't think incentive or disincentive to work is the main thing.

4/5ths of Americans with above $1 million income have business income. Actors, actresses, highly-paid sports players, and other celebrities may be the wealthy individuals seen most on TV, but they're not the majority.

Most small businesses and some larger ones are sole proprietorships, where the income of the business and the income of the owner are one and the same. Frequently most growth in the owner's income goes into growing the business. (While no doubt we've all heard of extravagant waste by some celebrities, often higher incomes gained by business founders correspond to personal consumption becoming a lesser fraction of the total; after all, nobody personally eats millions of dollars of food, for instance).

The matter is influenced by many forms of borrowing. Yet, fundamentally, the possible growth rate of a business is determined by its gain in revenue minus past expenses minus taxes, in other words by its aftertax profit, even if the motivation of the business owner was just as high either way. The business owner can't just count future new hirings of employees as a past expense to deduct from taxes for the past quarter, and, if he or she ends up with much more physical assets, those go into the calculation of taxable profit.

Of course, this is complicated by how a sole proprietorship can become a corporation to get taxed at the corporate tax rate instead, though not without tradeoffs, and, during times in past American history when the top individual income tax rate was at its highest, that was a common tactic, since the top corporate tax rate has never been above 50% and commonly been substantially less.

As a thought experiment, though, the dampening effect on business growth if tax rates were raised towards 100% is clear.

Isolating the effect of tax rates versus countless huge separate factors influencing growth would be very hard to do in an analysis. Still, a look at countries which started without exceptional natural resources per capita a few decades ago and yet managed to rapidly obtain high per capita wealth might be illuminating. Taiwan and Singapore, for instance, have low top tax rates for individuals and especially corporations.

(The trick, of course, is that must be combined with the rule of law, a government without too high levels of corruption, and a stable society, among other factors; Somalia, in contrast, is no haven for investment when the nearest warlord is liable to seize the business assets, and, in some impoverished countries, even land-line telephone networks have been held back by how often the copper wire gets looted).

Though not in that same group, China's top corporate tax rate appears to be only 25% while being up to 35% on individual income from a business (different from their 45% top rate on individual income from a salary). They started with a far higher tax rate or rather the 100% of pure communism not permitting such at all, but they got their exceptional growth since the 1980s by overcoming their natural starting ideology to make pragmatic decisions.

Obviously this isn't an argument for anarchy or for neglecting the often vital or useful role a number of government expenditures can provide, only something supportive of the TANSTAAFL principle when it comes to tax increases having some significant tradeoffs that must be weighed versus benefits. Those are well beyond mere "disincentive to work," applying even if motivation to work stays near-constant.

[Samuel]

The thing is the money doesn't just disappear- the government can use it to invest in infrastructure as well to boost growth. The question is how much of what is taxed would have been put to such a use and how much would have been spent in otherways.

The ratio between the two is significantly less clear.

[Master of Ossus]

A fair point. Perhaps a more thorough quoting of the original right-wing belief is in order: For example, "The Right has long held, and the Left is coming reluctantly to agree, that welfare creates a culture of dependency, sapping the initiative of its recipients. In the slums right now a generation of illegitimate children raised fatherless on Aid to Families with Dependent Children is being encouraged by welfare to produce the next generation." (from http://www.thefreemanonline.org/columns ... hould-end/ )

It has always been stated in such a manner as to strongly suggest that the "culture of dependency" effect can be generalized for the entire welfare population, which in turn implies that they refer to a majority, not just a barely measurable uptick in percentages.

It's tough to measure, but I've seen statistics holding that right when unemployment benefits cut off, there's something on the order of a 300% increase in the rate at which people get jobs, after which it (pretty quickly) drops back down. Because of this sort of study, though, the US has adopted Earned Income Tax Credit and basically modified it continuously ever since then because it's a compromise that truly satisfies nobody.

These would not be a true test of deterrence as they assume an equal criminality across all social groups. This is clearly not the case, crime seeming to be more a function of socio-economic factors and cultural factors (gangsta culture or corporate culture for example). A more proper empirical test would require two groups with similar social, economic and cultural backgrounds with one featuring the deterrent to be tested (and only that deterrent).

Testing for deterrence at all is very hard as well because it assumes that criminals at the time of their crime are rational actors, where as I am sure that it can be demonstrated via a quick perusal of statistics that a significant portion are under some intoxicant that would prevent them from fully considering the consequences of their actions and thus negating some of the deterrent effect that would be present in a normal individual. Which must also be taken into account when deciding whether a deterrent effect is actually functioning properly or not.

How does "testing for deterrence" assume "that criminals at the time of their crime are rational actors?" It just measures a correlation between an increase in punishment, or likelihood of being caught, or what have you and their observed behavior.

[Dark Hellion]

How does "testing for deterrence" assume "that criminals at the time of their crime are rational actors?" It just measures a correlation between an increase in punishment, or likelihood of being caught, or what have you and their observed behavior.

Deterrence only functions under the assumption that fear of punishment motivates the criminal to not commit the crime. Since we do not want to only look at correlations because of the complex underpinnings of crime, but we want to see directly whether increased punishment deters crime, we must take the psychological aspects of deterrence into account.

[Darth Wong]

How does "testing for deterrence" assume "that criminals at the time of their crime are rational actors?" It just measures a correlation between an increase in punishment, or likelihood of being caught, or what have you and their observed behavior.

Deterrence only functions under the assumption that fear of punishment motivates the criminal to not commit the crime. Since we do not want to only look at correlations because of the complex underpinnings of crime, but we want to see directly whether increased punishment deters crime, we must take the psychological aspects of deterrence into account.

No we don't. The whole idea of empirically testing things is that it's not enough to look at theoretically psychobabble, especially when the entire field of psychology is so historically unreliable. It's better to ask if there is empirical evidence to back up the cause-and-effect claims associated with tough sentencing. You don't appear to be getting this whole "look for empirical evidence" thing at all, because you're looking for excuses to examine these claims in strictly theoretical terms.

[Master of Ossus]

Deterrence only functions under the assumption that fear of punishment motivates the criminal to not commit the crime.

So under your assumptions, if there is a causal relationship between "fear of punishment" and a drop-off in crime, then deterrence is operating as it is supposed to. But your assumption is wrong. Looking for correlations doesn't assume any sort of theoretical underpinning. In fact, it is totally agnostic towards such things. After the correlation is observed, it may be possible to come up with theoretical mechanisms to describe the relationship (and it might even be possible to test those), but showing that there is a correlation does not require any philosophical framework.

Since we do not want to only look at correlations because of the complex underpinnings of crime, but we want to see directly whether increased punishment deters crime, we must take the psychological aspects of deterrence into account.

WHAT? Why do we not want to worry about the positive existence of correlations between some deterrent factor and crime rates? I would think we shouldn't care about the reasons for a drop-off, or increase, in crime if we can observe that this is affected by policy concerns. The point is to generate data to inform our policy decisions. Saying, "Not all murderers seem to be rational" does not do anything to inform policy considerations.

[Dark Hellion]

That was the point I was trying to make. Since crime is not a simple behavior, we cannot just look at regions with different punishment schemes and try to draw conclusion from them, as the causation of the crime may be different across different regions. So the only way to test for deterrent effect would either be to find two groups of incredibly similar peoples with different punishment schemes (highly unlikely) or figure out a way to empirically test the psychological effect of deterrence on the motivations to commit crime (again unlikely).

This is why it is very hard to test empirically whether or not punishment deters crime. We cannot establish a control nor can we find a testable mechanism from which to infer results.

So we really cannot say that it is a right-wing claim that is testable, it seems to be a matter of principal.

[Samuel]

You do realize that the punishments for crimes are occasionally changed? And that we can attempt to correct for differences in income, population and the like? This isn't an impossible task that you make it out to be- just hard.