Which Party Is Better for the Economy? by state
Guest post by Nathan Salminen (is a lawyer and an amateur economic and political researcher. Nathan has been politically active for many years, including working for the Senate Committee on the Judiciary as a law clerk. Nathan currently practices law in New York and runs Politics that work.)
Dan here…this is the second part to a series of three.
Comparing results between red and blue states
We have 50 different states, each with its own set of policies. In many cases, states have adhered to a generally consistent approach to policy for decades or even a century or more, so the condition the states are in is a strong indicator of the effects those policies have. This presents a great opportunity to evaluate the impacts of those policies by comparing the economies of the states.
The most straight-forward way to see whether the economy is performing better in red states or blue states is simply to look at the per-person median household income. The differences in that regard are stark:
You can see equally stark results by looking at the states ranked from highest to lowest for median household income.
Another way to measure which states are doing well economically is to look at the percentage of people in poverty. The results are the reverse of the results above for median income:
The success of the Democratic states seems to play out across the board, even beyond the economic realm, as is evidenced by the consistently stronger performance of the Democratic states in terms of standard of living. In fact, even lifespans differ dramatically between red and blue states.
It seems clear that liberal policies are performing better at the state level across the board. I posted some ideas about why Democratic states are doing so much better economically earlier, but for our purposes here all that matters is that they are in fact doing so much better.
It does not make any difference. The right relies on the ignorance and tribalism of the voters which the 1% manipulate to their advantage, nevermind that the statistics suggest that Elizabeth Warren is correct and everyone to the right of her–including most Democrats–are wrong. McConnell’s statement that the economy is improving because the GOP took both Houses of Congress was rated false today by the local fact checker, but 50% of the readers will figure that the fact checker is a liberal apologist. Ignorance of the GOP supporters and the apathy of the Democratic supporters, convinces me that capitalism is coming to an end and that we will be reverting to a kinder and gentler feudal society with a very small elite, a well educated and fairly prosperous service sector, a less well educated and less prosperous non service sector and a great mass of people living in poverty which will vary in how desperate it is by the wealth of the countries although not by as much as one would hope.
Liz Warren made a great presentation at the National summit on Raising Wages, put on by the AFL-CIO about policy choices with a history lesson first. https://www.youtube.com/watch?v=kjbiEo_sblU
Saw nothing about it on the bit of news I watched. In the past anytime the AFL-CIO did anything, it was on the news.
“The most straight-forward way to see whether the economy is performing better in red states or blue states is simply to look at the per-person median household income. ”
It may be straightforward, but it is somewhat sloppy. If costs are also lower (and they are) then income is not a good measure. A net value, or a change in income, or … would be better.
@Arne
Yes, cost of living is a tough one. In my view, the cost of living indexes are somewhat misleading. People in big cities trade off space inside their homes, for example, to get more access to museums, nightlife, interesting work, cuisine, diversity, etc. They are sacrificing tangible private resources for intangible public resources, but, cost of living calculators (as far as I have seen) only account for the tangible private resources. It is easy to compare the cost of a 3 bedroom, 2,000 square foot, apartment between two locations. It is also easy to control for things like AC, hardwood floors, etc., but it is not easy to control for things like one house has easy access to hiking while the other has easy access to museums.
Anyways, nonetheless, your point is totally valid and despite the above, there is some hard-to-define difference in cost of living. Good suggestion about change in income. I’ll give that a look. Thanks!
Arne/Nathan:
You may want to consider EPI’s 2013 study: http://www.epi.org/publication/ib368-basic-family-budgets/ “What Families Need to Get By”
@run75441/Arne
Thanks for the link. That is great data and I hadn’t seen it before. But, unfortunately, it looks at selected cities, so I can’t apply it to the states. That approach would be great though- subtracting the minimum to get by rather than adjusting the whole amount for COLA. The idea that COLA applies up to the getting by level, and that above that you get into inscrutable prioritization choices makes sense to me.
Maybe I have another approach. I could control for population density. At the most granular level I can find both COLA and median income data, just compare within brackets. Maybe I can do rural-to-rural and urban-to-urban or maybe it would need to be finer brackets. The problem with trading off between tangible private resources and intangible public resources, I think, mostly would occur between rural and urban living, so if I remove that from the equation, I’d be more comfortable applying a COLA.
I’ll take a look this weekend and post what I find back here, but please, let me know if any of you have any suggestions! It is very helpful to bounce this stuff off people with similar interests!
Last year the BEA started publishing state and metropolitan cost of living indicators for the 50 states and major cities. They publish if every quarter.
you can find it here:
http://www.bea.gov/newsreleases/regional/rpp/rpp_newsrelease.htm
Adjusting the nominal data for regional cost of living differences makes a few minor differences to the conclusion in this post, but the overall conclusion is still the same when you look at it in real terms.
Massachusetts is still about 125 % of the national average and Mississippi s still about 75% of the national average real per capita income.
Excellent, I’ll check that out as well. Thanks Spencer.
I posted a version of the same state median income chart adjusted for cost of living here- http://politicsthatwork.com/graphs/median-income-red-blue-states-rpp. Thank you all for the suggestions. The resulting graph is similar to the unadjusted one, although a bit less dramatic.
However, I am still not really convinced that just flatly adjusting state level data for cost of living is a particularly good way to go. I’m still mulling over the other ways to reflect those differences that we discussed, and I’ll post here if I come up with anything interesting.
I seriously question if there is anything to be gathered from these plots, other than the willingness of people to draw conclusions in agreement with their biases.
Correlation is not causation.
It is as likely that lower income people are more likely to be social conservatives, and thus the lower income causes the tendency to vote against the democrat politicians, as it is that somehow the policies of democrat politicians improve the lot of their constituencies. – This explains the move of the southern states away from the democrat party. The south would be happy to vote for democrats again, if the democrats hadn’t tied their standard to every imaginable form of evil, except perhaps bad economic policy.
While the republican party views these people with a combination of loathing and contempt, it is obvious that if they all stayed home every red state would turn blue.
It is easy for people to think everybody views the world the same as they do. But that isn’t really the case. While for some people, they really vote for people based on their best understanding of what is in their economic self interest, or what they think is sound or just economic policy, that is not the case for many people.
I know this is surprising, but many people vote for politicians based on their positions on social issues, or for other reasons (as an overwhelming percentage of ethnically African voters voted for the current occupant of the white house, presumably in part because of his ethnic origins)
And there are quite a lot of people who would accept poverty, to avoid the gross immorality which the democrat party advocates for.
So I suggest the conclusion that the economic policies of the democrats work better, while it may be true, cannot be demonstrated by these plots. I further suggest that if it were true, that many people wouldn’t consider this important enough that they would consider voting for a democrat.
@Dr. Jon
If the correlation only occurred at the state level, I might be inclined to agree that it could just be a propensity for social conservatism in poor states, but how do you explain that the economy also performs better at the national level when Democrats are in control? Part 3 may be forthcoming on this site, but you can read it now by following the link in this post if you like- it is the part about the national picture. IMO, the combination of the facts that blue states outperform red states and that the country performs better during Democratic years than during Republican years makes a pretty strong case that it isn’t just a confounding factor tied to either time or location.
At the state level, yes, social conservatism tends to correlate to poverty. But, I don’t think we can assume that is a one way street where poverty causes social conservatism. Certainly that happens- poorer people likely have less access to education and diversity, both of which tend to dispel socially conservative beliefs. But, the opposite happens too- social conservatism causes poverty. For example, two of the biggest economic boosts we’ve ever had in the US have come when we rolled back employment discrimination against women and minorities. Expanding the workforce and making employment decisions more on the basis of merit and less on the basis of irrelevant demographic factors drastically increased economic productivity. The same is likely going on now. For example, many socially conservative states still officially permit employment discrimination against gay people. If that in fact results in more of that kind of discrimination, that would be making people poorer in those states.
Thank you for your reply. I apologize for the delay in responding, but I have needed to deal with pressing personal matters, and this is the first opportunity I have had to respond.
I think you might have missed my main point, which is that it is difficult to tease out cause and effect from what amounts to epidemiological data. For an example (not wanting to discuss the unrelated issue, but picking an example from a totally unrelated field for example)
It is well documented that people who live in areas with high soil radon concentrations have a small but statistically significant lower rate of lung cancer than people living in areas where soil radon concentrations are significantly lower. There has been a lot of research by diverse organizations, but this is a consistent result. This is an interesting result, but I am still cautious regarding radon, even though the data shows it has a slightly protective effect. I am conflicted as to whether to believe the CDC, which says radon is the second largest cause of lung cancer, or the data gathered by government agencies and universities around the world, which tends to find no effect or even a slightly protective effect at exposure levels common in residential settings.
My point is that even large, well controlled studies with correction for confounding and well defined end points have difficulty getting meaningful results from this type of data.
My second point with regard to the social conservatism of the poor being due to their being in large more likely to be Christian in a meaningful sense than the rich has nothing to do with the complexity of their world, but is simply God’s preferential treatment of the poor.
When you suggest it is due to ignorance, you insult my family. While I am relatively educated and affluent, most of my extended family live in mobile homes, or houses they have built with their own hands. And while they lack formal education, I find that as a group, they and those I know in the communities where they live are not in any sense ignorant.
However, there is a correlation which is well documented, which is God’s outright preference for the poor. This is quite clear, and you can see it if you have eyes to see. This is also a common theme in the Bible, for example, “has not God chosen the poor to be rich in faith and to inherit the Kingdom he has prepared for those who love Him?”, and “He who is kind to the poor loans to God” – think about it, you can have the creator of the universe as your debtor just by showing kindness to the poor. And quite explicitly, “God chose the lowly things of this world, and the despised things – and the things that are not, to nullify the things that are, that no one may boast before Him.” This theme is repeated in many places, of which these are just a few examples. So it appears that the poor are socially conservative because of God’s preference for and choosing of them, not because they are too stupid to understand the issues of the day.
That is the correlation I was referring to, actually. Simone Weil referred to Christianity as the religion of slaves. To a large extent, she was correct.
Moving to the topic of which party is better for the economy, my current view is that neither party is good for the economy, and that if one is better than the other, it is impossible to demonstrate from the available data. That is more of a technical point. The reasons involve not just confusing correlation for causation, but also the problem of latency and the problem of confounding effects.
For an example of latency, consider that when I graduated High School, there was a lot of grant money available for low income students. This (as I understand it) was due to the Carter administration. As a result, I was able to get a pretty effective education, which increased my earnings throughout the Reagan and Clinton administrations. But the real economic payoff for Mr. Carter’s investment didn’t happen until the Bush administration, when my innovations began earning my employer in excess of $150,000,000 annually of added revenue. The resulting production lines employ many people in the US, both directly and indirectly. That added revenue stream continues to this day, without regard to the policies of the ensuing administrations. This is an anecdotal example, and shouldn’t be considered overly significant on a national scale, but it does illustrate the problem of latency pretty clearly, I think.
For an example of confounding, consider the current state of affairs in the Dakotas. The economy has boomed, due to technology advances which have made it economical to extract oil. It will likely bust now, due to the resulting crash in oil prices. A comparison of the performance of the economies of Minnesota to South Dakota makes no sense, unless the comparison correctly adjusts for this, which is very difficult to do.
Simple comparisons of cherry picked statistics seem to me to be meaningless and irresponsible, more suited to a political advertisement than a serious article.
Looking at the situation from a different perspective, it appears to me the primary issue we are facing, and have faced as a nation, is the debasing of the value of labor. The underlying assumption is that labor is a part of the cost of production, to be considered similarly to the cost of real estate, facilities, utilities, raw material, outside services, transportation, taxes, etc. It is a hard equation, where labor is viewed as a thing you buy, and it is virtuous to minimize what you pay for it.
This is a suppositional truth woven into the operational economic beliefs of people everywhere in the world. This is not everyone without exception, but everyone without distinction.
The primary way to debase the value of labor is to increase the supply, relative to the demand. The two primary ways this is being accomplished currently are to directly increase the labor pool, by unlimited immigration; and to indirectly increase the labor pool, by lowering tariffs so that labor from persons in foreign nations is not penalized. There is also the approach of mechanization or automation, but that is a somewhat smaller effect, though historically at times it has been dominant.
The effect of this on my family and friends is quite devastating. Real wages keep falling, without regard to which party is in power. Available employment opportunities are also restricted. It almost affected me directly also, as the job I had was moved to Beijing. I found other employment, but not everyone has the ability to do that, as my brother was pushed out of the job market at age 58 by similar forces.
Looking at the two political parties, it appears to me that both of them support these efforts to debase the cost of labor. While Republicans pretend to be opposed to illegal immigration, it is difficult to believe they are, when they held the White House for 8 years, and failed to stem the flood. Similarly, both parties support lowering of tariffs and eliminating restrictions which lower the cost effectiveness of offshoring.
While I would agree their formal positions are somewhat different, and the constituencies they claim to represent are significantly different, their behaviors are really quite similar. When you set aside the posturing, they are both the BIG Money party. If there are differences, they are subtle. I really don’t see them. I do agree that the Democrats have somewhat supported the union movement (more in the past than currently), but this seems to me not to be primarily due to a love for the working man, but to a love for the contributions of organized labor to their campaigns.
The extraordinary reaction against the public union breaking of Scott Walker seems to me to be a recognition that without the mandatory contributions which teachers make to the Democratic party (through their unions) the democratic party in Wisconsin will likely never regain power. It was a battle for survival for the Democrats. While they have done OK in Wisconsin since, it has largely been with money from outside of the state, and that is not sustainable long term.
Truly, there are individuals in both parties who are ideologues. Who actually believe the party line. I tend to view them favorably, even if I disagree with them.
But to establish the claim that one party or the other is “better” for the economy, first, you need to define what “better” is. What are the objective attributes? Real value of minimum wage? % of population earning minimum wage or less on an annualized basis? % of population at or below the poverty level? Workforce participation rate? Median real wealth (value of all the stuff a person has, minus liabilities)?
In establishing the criteria, it is necessary to do so before you look at the data. It is the cardinal sin of research to gather a bunch of data and look for a correlation. Because it is always possible to find a correlation which supports the premise “A is better than B” when looking at multifactoral data.
This is like someone adopting a particular view (like slavery is OK) and then going to the Bible to find justification. It is easy to find it, if you ignore the point of the text as a whole, and just look for proof texts. Alternatively, you can find that Charles Darwin was a racial supremacist, and an advocate for aggressive Eugenics, based on similar proof-texting from “The Descent of Man”, in which he discusses at large the lack of beneficial selection pressure on the human species. However, that isn’t a fair reading, as far as I can tell.
The temptation for this type of experimental error is great, and it corrupts a lot of research.
Next, you need to correct for latency. Latency effects in this case become really difficult to assess. For example enacting of laws which make it easier and more economically feasible for women with children to divorce will result in more divorces. This in turn will result in the impoverishment of many people, as the cost of maintaining two households where previously there was one will have a significant negative economic effect. I’m not arguing the morality of such legislation, only its practical effect, which I have experienced. It is a financial disaster. Many men are effectively forced out of the economy, as they would be required to pay 31% of their gross pay in child support, and cannot survive on what is left from the relatively modest paychecks they would otherwise receive. I’m not taking any side of this, but the situation of two households for what is one biological family is onerously expensive, and that is disproportionally so for the poor.
However, the effects are delayed. The immediate economic impact of legislation is minimal, and the effect continues indefinitely. There are also resulting changes in social values, which also have economic impacts as well. After correctly assessing the impact of such legislation, you would need to apportion responsibility for it to the political parties. And even in that case, the economic effects are not entirely negative, as being a single parent often motivates people to find ways of earning more money, getting additional education to earn more money, changing careers, etc. It also creates an entire industry of child care, not to mention the increased need for social services people and etc. This gets hard. Since every piece of legislation has an economic impact, every piece of legislation needs to be apportioned to the responsible party, and its economic effects, both direct and indirect, need to be assessed over time.
Next, you need to correct for confounding factors. These are things like natural disasters, or decisions of non-partisan organizations. This is relatively easy. How much of the current prosperity in Colorado (mostly Denver) is due to the run-up of oil prices, and the fact that the infrastructure support and many of the companies operating in the Dakotas are run out of Denver, and not out of St. Paul, MN? It is interesting that the oil boom has boosted the Colorado economy more than the Minnesota economy, though both have major population centers, and Minnesota is closer. But if you want to go state by state, you need to take those things into account. Also, the current unemployment rate in South Dakota is probably not due solely to it being a right to work state, but is to some significant extent due to the availability of employment in the oil exploration industry.
If you aren’t willing or able to do all of that work, then I think it is irresponsible to state that one or another party is better for the economy.
Overall the US political-economic situation looks like a parody of dialectical materialism to me. Democrats portray Thesis, Republicans portray Antithesis, and the resulting Synthesis is the debasement of the value of labor, which “benefits the economy”, because “It’s the economy, stupid!”
And of course, at the end of it all, if I was in charge I would need to consider what the actual ethics of employment are. And that gets complicated. Is it better to employ skilled labor in Cincinnati, at $21.50/hour, or to employ similarly skilled labor in Wuxi for the equivalent of $3.10 per hour? If we assume that labor is not a commodity, but has actual ethical content, we need to consider if it is better to provide employment to desperately poor people in Karnataka, or to relatively wealthy people in Philadelphia. Or are we to regard the prosperity which has been passed down to us from our parents a birthright? I would welcome a well-articulated discussion on that topic, though for now I will keep my tentative views to myself.
Pax.
Jon