Three Charts (not) to Email to Your Right-Wing Brother-In-Law

Truth-out.org suggests “Three Charts to Email to Your Right-Wing Brother-In-Law". Here are a few pictures to think about before you do so.

The graphs created by truth-out.org select the last year of the Bush presidency (not his entire term) to note that the recession started in 2008. Recessions actually happen very frequently in the US economy. This gives us a wealth of data with which to compare the current President’s performance. The first graph in the picture below plots the trends of many recessions. We evaluate using the same basis as truth-out.org--national employment statistics. In particular, we plot the employment as a percentage of its lowest value during a recession over time, for a number of recessions (from 1950 on). We can look at the growth of employment after the trough of a recession. Usually, the economy recovers rather quickly; in the graph, this corresponds to a steep uptick after hitting 0.

The current recession is the very bottom black line on the graph. This means that currently, we’re doing worse than any recession in the past 60 years in terms of recovery. Since Obama had been president for months before the trough, it wouldn’t seem natural to blame Bush alone for the slowest recovery that we have seen in the last 60 years. Just below the first graph, there is a comparison our recession--the blue line--with the average recovery after a recession. We are doing a poor job relative to the average.

In the second column, we look at another interesting trend: job growth before and after the passage of Obamacare. While the country seemed to be recovering at a reasonable rate and coming out of the recession before the passage of this bill, job growth slowed dramatically immediately following the passage of Obamacare in April 2010. Below this, the last graph depicts the actual stimulus rollout. The solid black line depicts funds received by organizations from the stimulus. Contrary to the deceptive picture painted by truth-out.org, the stimulus has been doled out at a fairly constant, increasing rate since the beginning of 2009. The article from truth-out.org, however, makes the ridiculous suggestion that the slowing of job growth and the constant unemployment rate is caused by the stimulus “winding down”. That is false. Just because there is no bill on the table doesn't mean that our economy isn't being "stimulated," and these are the results.

The Unemployed Held Hostage, Again

New York Times editorial "The Unemployed Held Hostage, Again" (November 27th, 2010) simply lies to its readers. It claims that there is no reason for tax cuts, but that extending unemployment insurance is a necessity. "There is no good argument for letting jobless benefits expire, or for extending those cuts."

Both of these claims are false. Harald Uhlig in an American Economic Review article "Some Fiscal Calculus" (also discussed here) estimates that for every dollar of government stimulus, in the long run time-discounted $3.40 is lost. For every dollar given up in taxation, time-discounted $2.40 is gained. On the other hand, Robert Shimer has estimated that unemployment insurance accounts for between 1-1.5% of the nation's unemployment rate.

Post-racial officials finding it's difficult to hold gains

Houston Chronicle article "Post-racial officials finding it's difficult to hold gains" (September 23rd, 2010) concludes that "black is black, white is white," and denotes the end of post-racial relations. Corrections suggests it gives little data for its conclusion.

A mere four years ago, for example, Americans celebrated the rise of two hope-filled post-racial stars: District of Columbia Mayor Adrian Fenty and Newark's Mayor Cory Booker. Each was touted as a young, energetic, post-civil-rights-era savior in the mold of then-Sen. Obama.

Multiracial coalitions of voters elected them both. Each managed to transcend race-based politics and make gains worthy of national praise regarding schools, crime, housing, economic developing and other issues that are either nonracial or should be.

Yet despite those successes, Fenty was voted out of office on Sept. 14. White voters still supported him almost two-to-one, but black voters in the majority-black city turned on him by a similarly wide margin.

What the article does not state is that Adrian Fenty lost to another black man, Vincent C. Gray. When one black man beats another black man in a Mayoral race, it's rather odd, without including more data, to conclude that skin color was a factor in the political race. Individual candidate race holds no predictive power in a simple regression (there is no variance in candidate race).

Holiday drivers will see a drop at the gas pump

Washington Post article "Holiday drivers will see a drop at the gas pump" (May 29th, 2010) makes a seemingly innocuous claim that doesn't appear to have empirical support. Specifically, the article claims that the approaching summer is a time in which gasoline prices increase.

Memorial Day weekend, the traditional start of the summer driving season, typically means higher prices at the gas pump. But drivers hitting the road for the holiday this year will be greeted with falling prices instead, thanks in part to investor jitters about the global economy that have sent crude oil prices tumbling in recent weeks.

The average monthly price data on regular unleaded gasoline does not appear to support the claim that "driving season" has meant higher prices historically. First, for reader edification, we display the inflation-adjusted relative price changes in regular unleaded gasoline since 1976, with the 1982-1984 price set to 1 (click to enlarge).



It is, of course, difficult to discern whether or not there is a modulo 12-month pattern from the time series display. Therefore, we produce the monthly deviations from a linear December-to-December trend (click to enlarge). If there is indeed a trend, we should see an increase during the Summer months, which certainly doesn't jump out at Corrections.



Viewing the data in a different way, we can see general monthly effects by looking at year-month price data (click to enlarge).




Finally, we can look at the same graph, but of off-December-to-December deviations (click to enlarge). If a trend is present, it should be apparent from a continual up-slope heading north-northwest on the figure between April and August--something that, if it is present, we cannot see.



In summary, a quick graphical examination of the data appears to belie the Post's statement that regular unleaded gasoline prices go up during the Summer. Corrections would further suggest that though we undoubtedly have a shift in the demand for gasoline, we also are likely to have a shift in supply. Indeed, we would suggest that one shouldn't expect a perpetual increase in price much above the cost for storing gasoline for a few months--otherwise a predictable and significant arbitrage opportunity is present--a thing abhorrent to nature.

The Magic Potion

New York Times editorial "The Magic Potion" (March 29th, 2010) appears to have no factual basis for its claim that poverty and homelessness are rising. Corrections conjectures they are falling, (we lack current data).

Poverty and homelessness are increasing.

Corrections has been unable to discover either current data or news articles on poverty or homelessness in 1Q 2010 (today being the second to last day, this is reasonable). We conjecture that Bob Herbert has made this current trend up, and look forward to posting data on this (just as we hope to do for the IHRSA , which has yet to post new data). We will use HUD's definition of homelessness and the census's definition of poverty, absent a change in how government publishes its statistics. We posit that both will be falling over first quarter 2010, at which Bob Herbert is writing at the tail end.

There is reason to believe we are correct. We graphically display GDP below up to and including the fourth quarter of 2009 (click to enlarge). We conjecture it will continue in much this trend. We will update this figure, along with homelessness and poverty statistics, when available.