DVD Talk
Bush Boom: Minority Report [Archive] - DVD Talk Forum

PDA

View Full Version : Bush Boom: Minority Report


bhk
10-14-05, 11:30 AM
http://www.nationalreview.com/nrof_buzzcharts/buzzcharts200510140859.asp

http://www.nationalreview.com/images/chart_bowyer10-14-05.gif

President Bush’s economic policies have created an environment that is highly conducive to the prosperity of racial minorities. BuzzCharts knows this assertion might shock some people, even those who are supporters of President George W. Bush. The Bush administration clearly is better at helping African Americans and Latinos become more prosperous than it is at telling the country what it has done to make these minority groups more prosperous.

BuzzCharts doesn’t do PR; we do data. And here’s what the data says:

During the last five years of President Clinton’s term — as far back as the data for this analysis goes — entrepreneurial activity among blacks increased 0.02 percent. Since the beginning of the Bush presidency, it has increased 0.04 percent — twice the Clinton growth rate for the available data. Meanwhile, entrepreneurial activity among Latinos increased 0.1 percent during the last five years of Clinton’s term. Since Bush has taken office, entrepreneurial activity among Latinos has jumped 0.12 percent.

As we have shown previously, black unemployment is near historic lows, as is the gap between black and white unemployment. So, as minorities searching for jobs are having greater success today, America is moving towards greater employment equality.

The chart above shows that this broad-based tendency towards prosperity is likely to continue. Business starts are a measure of dynamism in a community, and therefore of likely future prosperity. The new ownership society as envisioned by the current president is intended for everyone, and judging by the stats, the dream of business ownership is coming true for more blacks and Latinos under Bush than Clinton.

— Jerry Bowyer is the author of The Bush Boom and an economic advisor to Independence Portfolio Partners. He can be reached through www.BowyerMedia.com.



I don't expect to see this in the NYT for some reason.

kvrdave
10-14-05, 11:41 AM
They've done that in spite of Bush, no doubt. :lol:

I don't think there is anything that can legitimately be done with the perceptions that Black people have about Bush. Sadly, it probably just means having them blame others for their problems for eternity. :(

bhk
10-14-05, 11:42 AM
More entrepreneurial activity and less unemployment means at least those minorities aren't blaming others for their problems and are actually trying to get ahead.

sfsdfd
10-14-05, 11:45 AM
I don't expect to see this in the NYT for some reason.
During the last five years of President Clinton’s term — as far back as the data for this analysis goes — entrepreneurial activity among blacks increased 0.02 percent. Since the beginning of the Bush presidency, it has increased 0.04 percent — twice the Clinton growth rate for the available data.
Yeah, I don't understand why the Times hasn't issued a STOP THE HEADLINES order. For every 10,000 minority entrepreneurs we had in 2000, today we have 10,00<b>4</b>. Under Clinton, we would only have had 10,00<b>2</b>. George Bush is truly the best president ever!

- David Stein

X
10-14-05, 11:57 AM
I appears percentage increases in entrepreneurial activity among all races move in very increments so, even though a number like .04 appears small, a doubling from its previous .02 seems significant to me. Especially when compared with the extremely small increase among whites.

From what you hear you would think there would have been a decrease. We all know Bush isn't the first "Black President", his predecessor was. And his predecessor presided over a period of enormous entrepreneurial growth.

The latino increase is quite interesting.

sfsdfd
10-14-05, 12:01 PM
I appears percentage increases in entrepreneurial activity among all races move in very (small) increments so, even though a number like .02 appears small, a doubling seems significant to me.
"Very small increments" can easily be caused by random fluctuations - <i>especially</i> when we're talking about the economy. It's the macroeconomic equivalent of Brownian motion.

Is there any non-random-fluctuation explanation for this? Has there been a large push by the Bush administration to promote minority entrepreneurs? I must have missed that memo.

- David Stein

X
10-14-05, 12:03 PM
"Very small increments" can easily be caused by random fluctuations - <i>especially</i> when we're talking about the economy. It's the macroeconomic equivalent of Brownian motion.

Is there any non-random-fluctuation explanation for this? Has there been a large push by the Bush administration to promote minority entrepreneurs? I must have missed that memo.

- David SteinYou don't understand that when taxes are lowered and the business environment is improved all business activity goes up?

You should do the calculations to prove it is a random fluctuation if you want to claim that.

sfsdfd
10-14-05, 12:10 PM
You don't understand that when taxes are lowered and the business environment is improved all business activity goes up?
<i>All</i> business activity, sure. We're talking about <i>specific</i> business activity for minorities. I don't remember seeing a huge push by the administration to promote this segment, so crediting it with a data point cherry-picked from the ocean of macroeconometrics is specious.
You should do the calculations to prove it is a random fluctuation if you want to claim that.
It's an 0.02% gain. I think that speaks for itself.

- David Stein

X
10-14-05, 12:16 PM
<i>All</i> business activity, sure. We're talking about <i>specific</i> business activity for minorities. I don't remember seeing a huge push by the administration to promote this segment, so crediting it with a data point cherry-picked from the ocean of macroeconometrics is specious.I'd have to do some research on that. But maybe those initiatives weren't aimed at you so you didn't notice them.

However we have the resulting numbers that appear to show something happened among minorities that didn't happen among whites.

X
10-14-05, 12:18 PM
It's an 0.02% gain. I think that speaks for itself.

- David SteinI'm quite suprised that you made a statement like that.

kvrdave
10-14-05, 12:23 PM
<i>All</i> business activity, sure. We're talking about <i>specific</i> business activity for minorities. I don't remember seeing a huge push by the administration to promote this segment, so crediting it with a data point cherry-picked from the ocean of macroeconometrics is specious.

It's an 0.02% gain. I think that speaks for itself.

- David Stein

Ah, I think this is where the difference in thinking might be. It appears that you think that an increase in minority business would only happen as a result of a program targeted at minorities. I tend to think that anything that would promote business would tend to increase minorities, as they have a smaller percentage withing their population, thus it would gravitate up faster.

Sometimes minorities can improve themselves without a government program -wink-

usually it just involves government getting out of the way.

sfsdfd
10-14-05, 12:23 PM
I'm quite suprised that you made a statement like that.
Economic numbers don't have the precision of scientific measurements. If the trajectory of a probe in space spontaneously diverges by 0.02% - or even much less - that is significant and noteworthy, because there shouldn't be anything else affecting it.

Any useful number in economics is the result of a hundred different factors. For this reason, these useful numbers habitually exhibit random fluctuation. Thus, a small blip in either direction is not significant, and in fact is barely noticeable, against the background of fluctuation.

- David Stein

X
10-14-05, 12:32 PM
Economic situations are largely evaluated by their trends. Small numeric increases, particularly when they're related to very large numbers, such as the number of entrepreneurs, are completely valid measures. The numbers here appear to show trends that have more than a random meaning.

I ran into one initiative, I'm sure there are more. And there could also be the incentive to minorities of seeing the unprecedented number of blacks and latinos in top positions in this administration.

http://www.whitehouse.gov/news/releases/2004/07/20040723-5.html
http://www.whitehouse.gov/query.html?col=colpics&qt=minority

GFM
10-14-05, 12:35 PM
David, you're reading the graph incorrectly (and accompanying text is poorly written). It's not a .04% as opposed to .02% increase of the level of entrepreneurship. It's a .04% versus .02% increase of the total population entering in entrepreneurial ventures. When talking of a change from .30 to .34 of the total population, it's really a 13% increase in the overall level of black entrepreneurs.

http://www.kauffman.org/items.cfm?itemID=640

kvrdave
10-14-05, 12:40 PM
Probably just a fluctuation.

sfsdfd
10-14-05, 12:46 PM
The numbers here appear to show trends that have more than a random meaning.
I'd be pretty surprised if 0.02% is statistically significant. It may not even exceed the sampling error.
I ran into one initiative, I'm sure there are more.
Some of the cited initiatives obviously aren't going to have a visible impact yet (donating money to minority-heavy colleges.) Things like increasing minority business loans and granting tax credits to minority businesses may have impacted the data from the last year, but can't have heavily contributed to this data point cherry-picked over a five-year span.

In general, this just seems like a very desperate scrambling for a favorable metric of any kind. I'm sure that while Rome was burning under Nero's watch, the economy for water jugs experienced a trading uptick.

- David Stein

sfsdfd
10-14-05, 12:49 PM
It's not a .04% as opposed to .02% increase of the level of entrepreneurship. It's a .04% versus .02% increase of the total population entering in entrepreneurial ventures. When talking of a change from .30 to .34 of the total population, it's really a 13% increase in the overall level of black entrepreneurs.
I was going directly by the text of the article, which reads: "Entrepreneurial activity among blacks increased 0.02 percent." That is very different from: "Blacks gained 0.02% in their total representation among entrepreneurs." And the title of the graph doesn't much mitigate toward the latter interpretation.

- David Stein

hahn
10-14-05, 12:52 PM
David is completely right. I want to see a power analysis of this variation to see what the ability of this study is to capture a statistically significant difference. I also want to see 95% confidence limits of the data. Without this information, the data is completely useless and uninterpretable. Of course, leave it to people with very little understanding of statistics to interpet it the way they want to. We also need more context. Let's talk bottom line: has the overall amount of money generated by these minority entrepreneurs increased? If the number of entrepreneurs goes up, but overall, there is no gain or even a loss of revenue overall, would you blame Bush for not doing enough for minority businesses? I wouldn't. So why would I credit him for an increase?

Also, this argument about improved minority entrepreneurship would work much better IF this were one of the complaints of Bush's presidency. It's not. So I'm not sure how this argument is supposed to help prove that Bush is a good leader.

hahn
10-14-05, 12:54 PM
David, you're reading the graph incorrectly (and accompanying text is poorly written). It's not a .04% as opposed to .02% increase of the level of entrepreneurship. It's a .04% versus .02% increase of the total population entering in entrepreneurial ventures. When talking of a change from .30 to .34 of the total population, it's really a 13% increase in the overall level of black entrepreneurs.

http://www.kauffman.org/items.cfm?itemID=640

Which only marginalizes the actual increase in absolute terms, even more. This is a lame study with lame data that has absolutely no meaning at all without a further explanation of sample size, power, methodology, and error limits.

X
10-14-05, 12:57 PM
David, you're reading the graph incorrectly (and accompanying text is poorly written). It's not a .04% as opposed to .02% increase of the level of entrepreneurship. It's a .04% versus .02% increase of the total population entering in entrepreneurial ventures. When talking of a change from .30 to .34 of the total population, it's really a 13% increase in the overall level of black entrepreneurs.

http://www.kauffman.org/items.cfm?itemID=640Exactly.

A .0000002 increase in the growth of something is still a 100% increase over a .0000001 increase in growth and that's probably not an insignificant increase if that isn't a normally observed fluctuation.

The percent increase numbers may look small, but that's because they relate to very large numbers. So you look at the variations of the percent increases.

People who want to read the study instead of just argue can follow your link.

hahn
10-14-05, 12:59 PM
I was going directly by the text of the article, which reads: "Entrepreneurial activity among blacks increased 0.02 percent." That is very different from: "Blacks gained 0.02% in their total representation among entrepreneurs." And the title of the graph doesn't much mitigate toward the latter interpretation.

- David Stein

Yup. And what the EXACTLY does "entrepreneurial activity" mean? Someone submitted an application?

Let's put the numbers in another context assuming activity means business loan or registration of business. If during Clinton years, there were 10,000 applicants and it increased to 10,002. During the Bush years, it would have increased to 10,004. And this proves Bush's tax incentives are helping the economy? You have GOT to be kidding me.

X
10-14-05, 01:02 PM
I was going directly by the text of the article, which reads: "Entrepreneurial activity among blacks increased 0.02 percent." That is very different from: "Blacks gained 0.02% in their total representation among entrepreneurs." And the title of the graph doesn't much mitigate toward the latter interpretation.

- David SteinYou missed the part where it said "twice the Clinton growth rate for the available data."?

hahn
10-14-05, 01:05 PM
David, you're reading the graph incorrectly (and accompanying text is poorly written). It's not a .04% as opposed to .02% increase of the level of entrepreneurship. It's a .04% versus .02% increase of the total population entering in entrepreneurial ventures. When talking of a change from .30 to .34 of the total population, it's really a 13% increase in the overall level of black entrepreneurs.

http://www.kauffman.org/items.cfm?itemID=640

I don't see where David misinterpreted this. We understand that we are talking about percentage increase. The problem is that (as far as I read), there's no indication what the absolute numbers of minority entrepreneurs there are or what proportion of the overall entrepreneurs they comprise. Let's say there are 100,000 entrepreneurs. If half of them are minorities (50,000), a 1% increase (+500) is much larger than if only 1% of the entrepreneurs were minorities (1,000) with the same 1% increase (+10).

GFM
10-14-05, 01:07 PM
Yup. And what the EXACTLY does "entrepreneurial activity" mean? Someone submitted an application?

Let's put the numbers in another context assuming activity means business loan or registration of business. If during Clinton years, there were 10,000 applicants and it increased to 10,002. During the Bush years, it would have increased to 10,004. And this proves Bush's tax incentives are helping the economy? You have GOT to be kidding me.

ok, let's work with more or less real numbers. From the index (read the link I put up earlier for methodology), % of black entrepreneurs

1996 - .29
2000 - .31
2005 - .35

Let's assume a population of 300,000,000 for the US (easy to work with) and a black population of 12%. We can roughly say there are 36,000,000 black people in the US. What the index is saying is

in 1996 there were approximately 104,400 black entrepreneurs
in 2000 there were approximately 111,600 black entrepreneurs
in 2004 there were approximately 126,000 black entrepreneurs

These are not statistically insignificant

sfsdfd
10-14-05, 01:08 PM
From the study text:
<b>It is too early, however, to know whether the recent increase in the entrepreneurship is due to cyclical or structural factors.</b>
Even the study acknowledges that it's not known why this occurs. "Cyclic" and "structural" factors are the causes of the random fluctuation in these numbers.

As for margin of error: Figure 1 from the Report is pretty damning...
<img src="http://www.djstein.com/Misc/EconomicReportFigure1.jpg">
Figure 3, which dices up the figures by season, is even more chaotic.

- David Stein

hahn
10-14-05, 01:09 PM
Another fault of the study is that the data only goes back for the last 5 years of Clinton's terms. Well, if the majority of the increase took place in the first 3 years (1992-1995), then the data would completely off. Also another reason why the interpretation and conclusions of this data is complete bunk. This guy's "statistical analysis" is complete joke.

GFM
10-14-05, 01:09 PM
From the link earlier. Btw, number I put up should be revised down by some amount (missed the part about adult population)

The Kauffman Index of Entrepreneurial Activity finds that over the period from 1996 to 2004, an average of 0.36 percent of the adult population created a new business each month, representing approximately 550,000 new businesses per month. The rate of overall entrepreneurship activity remained relatively constant over the period despite major changes in the economy, with the rate of business creation generally between 0.3 and 0.4 percent. The average rate of entrepreneurship was 0.36 percent in 1996, 0.35 percent in 2001, and rose to 0.4 percent in 2004. It is too early, however, to know whether the recent increase in the entrepreneurship is due to cyclical or structural factors.

hahn
10-14-05, 01:18 PM
You also need to address "hypothesis plausibility". How many enterpreneurial start-ups made the decision to start-up based on Bush's tax incentives? IOW, without them, how many would NOT have gone ahead with a start-up? I'm planning on creating a startup myself next year. It has absolutely nothing to do with tax incentives or Bush. In fact, I have no idea how much money I'd be saving with Bush over someone else. Do you? If you don't, then how do you create the hypothesis that a tax incentive is the reason for more people wanting to create a start-up.

sfsdfd
10-14-05, 02:17 PM
Here are some other, um, telling figures:

<img src="http://www.djstein.com/Misc/EconomicReportFigure3.jpg">
I think I see a trend in there somewhere, if I squint a little and tilt my head...

<img src="http://www.djstein.com/Misc/EconomicReportAppendixFigure1.jpg">
It's interesting that this minuscule blip is visible only in one of the several common measurements of "entrepreneurship." By another measure (low-hours business ownership), it's actually gone down.

I think we're done here.

- David Stein

shifrbv
10-14-05, 02:27 PM
How many enterpreneurial start-ups made the decision to start-up based on Bush's tax incentives? IOW, without them, how many would NOT have gone ahead with a start-up?

I think Bush's tax incentives have very little to do with these entrepeneurs decisions to open their own business. I would like to see the data on the number of people who chose to open their own business because they couldn't get hired anywhere else. For latinos, this is especially true and would explain the increase.

Also, I think it would be interesting to see if these people are financially better off. Competition is very tight, and I would be surprised if these people were able improve their standard of living. The days of making good profit in "small business" are pretty much over.

X
10-14-05, 02:35 PM
I think we're done here.I think you should look at the numbers of people shown in the study. And my original take of this thread was that Bush hasn't been toxic to black success, particularly when compared with the president who was praised for being so good to them. That hypothesis has not been disproven. In fact, if you look at measures such as black home ownership, a case could be made for that things have improved for blacks under Bush's administration.

sfsdfd
10-14-05, 02:48 PM
I think you should look at the numbers of people shown in the study.
If the percentage change is within normal variance - as, hopefully you'll now agree, it clearly is - then the total number over which that percentage occurs is irrelevant. It's just part of the random walk. It's not indicative of a trend.

Even the study plainly admits that.
And my original take of this thread was that Bush hasn't been toxic to black success, particularly when compared with the president who was praised for being so good to them.
I agree that Bush hasn't been "toxic" to black success. My impression had been that he's been pretty hands-off. The recent efforts listed in your link show that Bush has more recently taken up this cause with effective measures, and that's a good thing; but it's not logical to cite this data as evidence in his favor (as the OP tried to do without applying any critical analysis.)

- David Stein

X
10-14-05, 02:55 PM
If the percentage change is within normal variance - as, hopefully you'll now agree, it clearly is - then the total number over which that percentage occurs is irrelevant. It's just part of the random walk. It's not indicative of a trend.I think any individual percentage change can be shown to be within an error margin. I'm not convinced the overall trend can be show to be due to error.

Ranger
10-14-05, 03:11 PM
I don't trust government data, and I certainly don't trust data from the National Review. But even if this trend is a fact, anyone who seriously believes that Bush and his policies alone were responsible for it is rather disillusioned.

And why only look at this trend? the trend of increasing poverty should be a much bigger issue.

Tommy Ceez
10-14-05, 03:14 PM
I don't trust government data, and I certainly don't trust data from the National Review. But even if this trend is a fact, anyone who seriously believes that Bush and his policies alone were responsible for it is rather disillusioned.

And why only look at this trend? the trend of increasing poverty should be a much bigger issue.
This post perfectly summs up this entire thread.

hahn
10-14-05, 06:18 PM
You don't understand that when taxes are lowered and the business environment is improved all business activity goes up?

You should do the calculations to prove it is a random fluctuation if you want to claim that.
So you've drawn a conclusion and because PART of the data doesn't go against that conclusion, then it proves the conclusion must be true? :hscratch:

And no, we don't need to do calculations to prove a random fluctuation. You don't prove randomness in statistics. You prove trends or statistical significance. It's random until proven otherwise, not the other way around.

hahn
10-14-05, 06:26 PM
ok, let's work with more or less real numbers. From the index (read the link I put up earlier for methodology), % of black entrepreneurs

1996 - .29
2000 - .31
2005 - .35

Let's assume a population of 300,000,000 for the US (easy to work with) and a black population of 12%. We can roughly say there are 36,000,000 black people in the US. What the index is saying is

in 1996 there were approximately 104,400 black entrepreneurs
in 2000 there were approximately 111,600 black entrepreneurs
in 2004 there were approximately 126,000 black entrepreneurs

These are not statistically insignificant
#1 That's not what determines "statistical significance".
#2 Why do you conclude that if there is a trend of increasing black entrpreneurs that it must be due to Bush? If I drink some soda right now and tomorrow, I wake up with a headache, does that mean the soda caused my headache? You're jumping to a causal relationship without sufficient plausibility OR statistically significant data to support the OP's hypothesis: that Bush's economic policies have fostered an increase in minority entrepreneurship.

DVD Polizei
10-14-05, 06:55 PM
I agree with David. And I hope that's not a threadcrap.

GFM
10-14-05, 07:39 PM
#1 That's not what determines "statistical significance".
#2 Why do you conclude that if there is a trend of increasing black entrpreneurs that it must be due to Bush? If I drink some soda right now and tomorrow, I wake up with a headache, does that mean the soda caused my headache? You're jumping to a causal relationship without sufficient plausibility OR statistically significant data to support the OP's hypothesis: that Bush's economic policies have fostered an increase in minority entrepreneurship.

Looking at the methodology behind the study, it looks like the sample size is absolutely enough to declare it as a significant population and there looks to be a yearly upward trend (with an expected blip post dot.com). What are you saying makes it insignificant? It's been 10 years since my last stats class, so maybe I'm missing something.

I'm not making any conclusion that is was due to Bush. I'm merely saying it's an increase that's not statistically insignificant. Point out where I wrote it was due to Bush or made any relationship to causal effect. I'm merely presenting data and presenting how these numbers work out more or less in the real world; you're attributing things I have never written to me.

X
10-14-05, 08:02 PM
I agree with David. And I hope that's not a threadcrap.I hope you can tell the difference, but I'm glad you got the message.

hahn
10-15-05, 01:55 AM
Looking at the methodology behind the study, it looks like the sample size is absolutely enough to declare it as a significant population and there looks to be a yearly upward trend (with an expected blip post dot.com). What are you saying makes it insignificant? It's been 10 years since my last stats class, so maybe I'm missing something.

I'm not making any conclusion that is was due to Bush. I'm merely saying it's an increase that's not statistically insignificant. Point out where I wrote it was due to Bush or made any relationship to causal effect. I'm merely presenting data and presenting how these numbers work out more or less in the real world; you're attributing things I have never written to me.

I assumed you were in support of the implication made by the OP since you were defending the data, and he was using the data to support Bush as a president. The implication by the OP was quite clear. If you disagreed with the validity of that interpretation of the data, then I apologize for unfairly attacking your assessment.

In regards to statistical significance, it is not determined by sample size. Sample size determines power - the ability to detect Type I error. In simple terms, it determines your ability to detect a pattern if it's present in a given set of data. Statistical significance has to do with the chance of whether your establishment of a pattern is in fact real, or whether you found a difference by mere chance. Like flipping a coin 100 times. You're supposed to get 50 heads and 50 tails. But by chance, you could also get 65 heads and 35 tails. It doesn't mean that you really have a better chance of getting heads than tails.

Statistical signficance has traditionally been defined and accepted by statisticians as 1.98 times the standard deviation of a data set. It establishes a range of data which, if a given data point falls within that range, then there is a 95% chance that your data was not due to random chance. So if you were to do multiple 100 flip sets of a coin, and you had an average of 50 heads and 50 tails after, say 50 sets of this (5000 coin flips), and you find that the standard deviation is, let's say, 10. Then your 95% Confidence Interval range would be 30.2-69.8 (you add and subtract 1.98*10 from the average). What this means is: if you do another set of 100 coin flips and you get 75 heads, there is only a 5% chance that your result was due to random chance. There's a 95% chance that there was some other "thing" that influenced the coin toss to favor getting 75 heads. The difference between your result and the reference result is statistically signficant. See?

Trend is also a distinct term. When we're talking about trends, it's not only important to establish power, and statistical signficance. We also need to create a meaningful model; a set of variables which creates a theoretical pattern which we need to follow over time. Like determining changing seasonal weather patterns by figuring out the determining factors which we think contribute to the weather and then following those factors over time, and determining (mathematically) the validity of your model. And as I said, even if there were a trend, this does nothing to estabilish causality (in this case the proposed causality of Bush's tax incentives causing minorities to increase entrepreneurial "activity".

Like I said, the link you presented, says absolutely NOTHING about any of the above. Also, I realize that your link only presents the data. But the OP posted a link of someone who decided to present a political interpretation of the data. Followed up by his own editorial sneer of how this wouldn't be published in the NYT. Of course it wouldn't. It's a moronic interpretation.

classicman2
10-15-05, 08:09 AM
Bush Boom: Consumer prices rose in September at the highest rate in 25 years.

Reminds me of the Carter days.


Content Relevant URLs by vBSEO 3.2.0