In a recent post, I discussed a survey that found that liberals were a little more likely to say that conservatives shared their values and goals than conservatives were to say that about li erals. However, the difference wasn't statistically significant. It turns out that that same question was asked in a survey a month later. In this survey, the differences were stronger and significant. 44% of conservatives said that most liberals shared their values, while 56% of liberals said most conservatives shared their values. (The difference is a little bigger--42% to 61%--if you use the sampling weights).
The survey also asked about patriotism. Only 4% of conservatives thought that liberals were usually more patriotic than conservatives, while 29% said they were less patriotic (61% said the same, ad the rest said they didn't know). Liberals overwhelmingly (81%) said there was no difference: 10% of liberals said that conservatives were more patriotic, 6% said liberals were more patriotic.
Thursday, February 7, 2013
Saturday, February 2, 2013
People today
A few weeks ago, I had a post about a series of questions on whether "young people today" had as strong a sense of right and wrong today as they used to. There has also been a question on "people in general": specifically "Do you think people in general today lead as good lives--honest and moral--as they used to?" The responses, as percent saying yes minus percent saying no:
Views have become much more negative: from evenly split (47% yes, 46% no) in 1952 to 21% yes and 74% no in 2005. I'm not surprised that opinions became more negative between the 1950s and the 1970s, but it's interesting that they've continued to become more negative since the 1970s. Even by most traditional standards of morality (crime, divorce, drug use), Americans have generally improved since the 1970s.
Views have become much more negative: from evenly split (47% yes, 46% no) in 1952 to 21% yes and 74% no in 2005. I'm not surprised that opinions became more negative between the 1950s and the 1970s, but it's interesting that they've continued to become more negative since the 1970s. Even by most traditional standards of morality (crime, divorce, drug use), Americans have generally improved since the 1970s.
Thursday, January 31, 2013
Is anything else holding back the recovery?
In a recent post, I compared nations and found no evidence that ones with more inequality are recovering more slowly from the recession. That led me to think about other ideas of what might hold back the recovery. One thing that's sometimes said is that measures which make wages less "flexible" make it harder to rebound from a recession. This argument relies on the basic economic principle that if the price of something becomes lower, people will buy more of it. So if wages decline during a recession, employers will buy (hire) more. If they don't, that's good for people who still have jobs, but employers won't hire, and unemployment will remain high. A paper by Juan Botero et al. compiled measures of the strength of employment protection laws, collective bargaining laws, and social security measures in different countries.
Employment protection includes things like regulation of work hours and laws limiting employers' ability to lay off workers. Collective bargaining laws protect workers' right to unionize and encourage or require employers to negotiate with unions. "Social security" includes unemployment insurance and benefits for retirement, disability, and sickness. The measures are available for 30 of the 33 OECD countries (all except Estonia, Iceland, and Luxembourg). The data are from about 2000, but things like that usually don't change rapidly, so they're probably still pretty accurate for the recession.
If you regress economic growth from 2009-11 on these factors plus log GDP and stimulus spending, you get:
estimate t-ratio
Constant .702 2.82
Wealth -.058 -2.23
Stimulus .54 2.25
Employment -.046 -1.97
Bargaining -.001 -.01
Security -.021 -.71
That is, there's no evidence that social security or protection for collective bargaining makes a difference, but some evidence that employment protection does. If we drop social security and protection for collective bargaining, we get:
estimate t-ratio
Constant .759 3.26
Wealth -.067 -3.09
Stimulus .556 2.32
Employment -.049 -2.67
The US ranks low in employment protection, but so do some countries with generous welfare states, like Sweden and Denmark. The countries that rank high in employment protection are mostly in southern Europe (Portugal, Spain, Greece) or eastern Europe (Poland, Slovak Republic, Slovenia).
In terms of the combination of policies that's associated with faster recovery (high stimulus spending, low employment regulation), South Korea ranks at the top, followed by Austria, Canada, and the United States. At the other extreme is Portugal, followed by Hungary, and Greece.
Employment protection includes things like regulation of work hours and laws limiting employers' ability to lay off workers. Collective bargaining laws protect workers' right to unionize and encourage or require employers to negotiate with unions. "Social security" includes unemployment insurance and benefits for retirement, disability, and sickness. The measures are available for 30 of the 33 OECD countries (all except Estonia, Iceland, and Luxembourg). The data are from about 2000, but things like that usually don't change rapidly, so they're probably still pretty accurate for the recession.
If you regress economic growth from 2009-11 on these factors plus log GDP and stimulus spending, you get:
estimate t-ratio
Constant .702 2.82
Wealth -.058 -2.23
Stimulus .54 2.25
Employment -.046 -1.97
Bargaining -.001 -.01
Security -.021 -.71
That is, there's no evidence that social security or protection for collective bargaining makes a difference, but some evidence that employment protection does. If we drop social security and protection for collective bargaining, we get:
estimate t-ratio
Constant .759 3.26
Wealth -.067 -3.09
Stimulus .556 2.32
Employment -.049 -2.67
The US ranks low in employment protection, but so do some countries with generous welfare states, like Sweden and Denmark. The countries that rank high in employment protection are mostly in southern Europe (Portugal, Spain, Greece) or eastern Europe (Poland, Slovak Republic, Slovenia).
In terms of the combination of policies that's associated with faster recovery (high stimulus spending, low employment regulation), South Korea ranks at the top, followed by Austria, Canada, and the United States. At the other extreme is Portugal, followed by Hungary, and Greece.
Sunday, January 27, 2013
Congress, Cockroaches, and France
A recent survey that's been getting some attention asks people whether they have a more favorable opinion of Congress or various widely disliked things. Congress ranks behind many of them--cockroaches seem to be getting the most publicity. On looking at the complete list, I noticed one that seemed out of place--France. France did get more approval than Congress (46%-37%), but the margin wasn't impressive: traffic jams got more approval by 56%-34%. Do Americans really dislike France that much?
I found a number of questions of the basic form "What is your overall opinion of France? Is it very favorable, mostly favorable, mostly unfavorable, or very unfavorable?" A few of them just asked whether it was favorable or unfavorable, so I collapsed them all into favorable and unfavorable and took the difference between favorable and unfavorable ratings.
Favorable opinions have consistently outnumbered unfavorable ones, except in 2003-4. That stretch of unpopularity was presumably the result of France's opposition to the Iraq war, but apparently most of us have forgiven them. The latest survey (Feb 2012) had 75% favorable and 21% unfavorable, which is somewhat below the favorability ratings in the 1990s, but still pretty good.
I would try to figure out a way to reconcile the results, but the survey on Congress was an automated one (press 1 for .... ), and as I've said before, I suspect that people don't take automated surveys very seriously. I hope they don't--otherwise we have to conclude that Americans are pretty fond of lice, which beat Congress by 67%-19%.
I found a number of questions of the basic form "What is your overall opinion of France? Is it very favorable, mostly favorable, mostly unfavorable, or very unfavorable?" A few of them just asked whether it was favorable or unfavorable, so I collapsed them all into favorable and unfavorable and took the difference between favorable and unfavorable ratings.
Favorable opinions have consistently outnumbered unfavorable ones, except in 2003-4. That stretch of unpopularity was presumably the result of France's opposition to the Iraq war, but apparently most of us have forgiven them. The latest survey (Feb 2012) had 75% favorable and 21% unfavorable, which is somewhat below the favorability ratings in the 1990s, but still pretty good.
I would try to figure out a way to reconcile the results, but the survey on Congress was an automated one (press 1 for .... ), and as I've said before, I suspect that people don't take automated surveys very seriously. I hope they don't--otherwise we have to conclude that Americans are pretty fond of lice, which beat Congress by 67%-19%.
Tuesday, January 22, 2013
Is Inequality Holding back the Recovery?
In the New York Times, Joseph Stiglitz says "Inequality is Holding Back the Recovery," but Paul Krugman is skeptical and says there no reason to expect that inequality will affect short-term economic growth. Using data from the Organisation for Economic Cooperation and Development, here is the relationship between inequality (the Gini coefficient) as of about 2005 and economic growth between 2009 and 2011--that is, since the bottom of the recession.
It looks like the more unequal countries have recovered more rapidly. But most of the countries that combine strong growth and high inequality are also poor (relative to the OECD nations as a whole), so they have more room to grow. As I said in an earlier post, the amount of stimulus (government spending growth) in 2005-7 also seems to have affected economic growth in 2009-11. If you use all three of these factors to predict growth, you get
t-ratio
Gini .01 0.1
Stimulus .34 1.4
wealth -.05 -2.2
Of course, that's not proof that inequality makes no difference: the confidence interval (range of plausible values) is large enough to include pretty substantial effects, both negative and positive. But there's no sign that countries with more inequality have had a slower recovery from the recessions. So this evidence comes down on Krugman's side.
Notes: "Wealth" is the logarithm of per-capita GDP, as of 2009. For "stimulus," see my earlier post on the subject. Greece is an outlier on economic growth, but it doesn't make much difference to the results, since it's pretty much average in terms of the three predictors.
t-ratio
Gini .01 0.1
Stimulus .34 1.4
wealth -.05 -2.2
Of course, that's not proof that inequality makes no difference: the confidence interval (range of plausible values) is large enough to include pretty substantial effects, both negative and positive. But there's no sign that countries with more inequality have had a slower recovery from the recessions. So this evidence comes down on Krugman's side.
Notes: "Wealth" is the logarithm of per-capita GDP, as of 2009. For "stimulus," see my earlier post on the subject. Greece is an outlier on economic growth, but it doesn't make much difference to the results, since it's pretty much average in terms of the three predictors.
Friday, January 18, 2013
Liberals according to conservatives, conservatives according to liberals
A 2006 CBS News surveys asked people who said they were conservative a few questions about liberals and asked people who said they were liberals the same questions about conservatives (people who said they were moderate weren't asked about either). The questions were:
1. People who are ____ feel differently than I do about politics, but they probably share many of my other values and goals; OR 2. People who are _____ feel differently than I do about politics, and they probably do not share many of my other values or goals, either.
2. In general, do you think most _____ have beliefs and ideas that are
usually reasonable, even if you disagree with them, or are their beliefs and
ideas usually unreasonable?
3. Compared to most [your group], do you think most [other group] generally
tend to be richer, poorer, or economically about the same as most [your group]?
4. Compared to most [your group], do you think most [other group] are
generally more informed about the world, less informed, or about as
informed about the world as most [your group]?
The following numbers show what each group says about the "other side."
Conservatives Liberals
Share values 39% 44%
Don't share 57% 54%
Reasonable 45% 51%
Unreasonable 48% 41%
Richer 26% 58%
Same 57% 36%
Poorer 9% 2%
More informed 9% 6%
Same 67% 58%
Less informed 21% 32%
For example, 39% of conservatives say that liberals share their values and goals; 44% of liberals say conservatives share their values and goals. On the first two questions, liberals seem to have a slightly more favorable view of conservatives than conservatives do of liberals (the differences aren't statistically significant, but are close enough to be worth taking seriously). On the third, there's a definite tendency to see "them" as richer than "us," but it's stronger among liberals--most conservatives think that the two groups are the same, while most liberals think that conservatives are richer. On the last, there's a tendency to see "them" as less informed, and it is stronger among liberals.
To summarize, it seems like conservatives are (a little) more likely to think of liberals as unreasonable people who don't share their values, and liberals are more likely to think of conservatives as poorly informed.
1. People who are ____ feel differently than I do about politics, but they probably share many of my other values and goals; OR 2. People who are _____ feel differently than I do about politics, and they probably do not share many of my other values or goals, either.
2. In general, do you think most _____ have beliefs and ideas that are
usually reasonable, even if you disagree with them, or are their beliefs and
ideas usually unreasonable?
3. Compared to most [your group], do you think most [other group] generally
tend to be richer, poorer, or economically about the same as most [your group]?
4. Compared to most [your group], do you think most [other group] are
generally more informed about the world, less informed, or about as
informed about the world as most [your group]?
The following numbers show what each group says about the "other side."
Conservatives Liberals
Share values 39% 44%
Don't share 57% 54%
Reasonable 45% 51%
Unreasonable 48% 41%
Richer 26% 58%
Same 57% 36%
Poorer 9% 2%
More informed 9% 6%
Same 67% 58%
Less informed 21% 32%
For example, 39% of conservatives say that liberals share their values and goals; 44% of liberals say conservatives share their values and goals. On the first two questions, liberals seem to have a slightly more favorable view of conservatives than conservatives do of liberals (the differences aren't statistically significant, but are close enough to be worth taking seriously). On the third, there's a definite tendency to see "them" as richer than "us," but it's stronger among liberals--most conservatives think that the two groups are the same, while most liberals think that conservatives are richer. On the last, there's a tendency to see "them" as less informed, and it is stronger among liberals.
To summarize, it seems like conservatives are (a little) more likely to think of liberals as unreasonable people who don't share their values, and liberals are more likely to think of conservatives as poorly informed.
Sunday, January 13, 2013
Labor Shares Around the World
A few weeks ago, Paul Krugman noted that the share of compensation in national income has declined over the last 40 years or so. That is, labor (broadly defined as everyone who gets a wage or salary) is relatively getting less and capital is getting relatively more. Thomas Edsall also wrote about the issue last week. I wondered whether this was happening in other countries too. Here are figures on labor share based on OECD data from 1975 until recently (between 2009 and 2011 depending on the country).
Trend 1975 Latest
1 UK -.037 70.1% 70.7%
2 US -.099 68.3% 63.7%
3 Canada -.161 68.1% 59.7%
4 Australia -.172 66.9% 55.9%
5 Germany -.219 74.5% 68.3%
6 Sweden -.241 72.9% 63.5%
7 France -.350 75.4% 68.9%
8 Japan -.394 68.0% 59.4%
9 Italy -.494 82.6% 68.0%
The "Trend" is the average change per year in that country. For example, in the US, the average annual change is 0.1%, which means 3.5%-4% over 35 to 40 years. The trend is negative in every nation, but its size varies greatly, and the decline in the United States is relatively small compared to other countries. Looking at it another way, in 1975 the labor share in the US was eighth out of nine countries (just ahead of Canada); in the latest year the US ranked fifth.
PS: I limited this to countries that were relatively affluent by world standards at the beginning of the period. Labor share seems to fluctuate more from year to year in small countries, so I included only a few: Canada and Australia because they are culturally similar, and Sweden because sociologists always like to compare the US and Sweden.
Trend 1975 Latest
1 UK -.037 70.1% 70.7%
2 US -.099 68.3% 63.7%
3 Canada -.161 68.1% 59.7%
4 Australia -.172 66.9% 55.9%
5 Germany -.219 74.5% 68.3%
6 Sweden -.241 72.9% 63.5%
7 France -.350 75.4% 68.9%
8 Japan -.394 68.0% 59.4%
9 Italy -.494 82.6% 68.0%
The "Trend" is the average change per year in that country. For example, in the US, the average annual change is 0.1%, which means 3.5%-4% over 35 to 40 years. The trend is negative in every nation, but its size varies greatly, and the decline in the United States is relatively small compared to other countries. Looking at it another way, in 1975 the labor share in the US was eighth out of nine countries (just ahead of Canada); in the latest year the US ranked fifth.
PS: I limited this to countries that were relatively affluent by world standards at the beginning of the period. Labor share seems to fluctuate more from year to year in small countries, so I included only a few: Canada and Australia because they are culturally similar, and Sweden because sociologists always like to compare the US and Sweden.
Subscribe to:
Posts (Atom)