President Trump?

[Loyal Rogue]

Ok. Since you insisted, I went through the Saez paper, looking for explanations of the points I listed.

There is only one that I see being addressed at all: changes in household size.

Saez argues that household changes were uniform between income groups. He offers no source or data to support this assumption. So, you will need to supply that if you believe it to be true.

Furthermore, even if household changes over time were uniform between income groups, that doesn't mean that factoring them in wouldn't significantly change the data. In fact, it would cause all income groups to show more growth.

Not only did Saez not address the issues you claim he does, he confirms that the problems are in the data when describing his methodology. He specifically lists the things that are counted as income ("salaries and wages, dividends, interest income, rents and royalties, and business income") and they do not include entitlements and benefits. He removes a massive amount of income that disproportionately calculates into the lower groups incomes.

This is a classic case of garbage in, garbage out statistics.

Fair enough.
I don't dispute that garbage in does equal garbage out.

As Burtless pointed out his his assessment of the debate, both parties are using different sets of data each with their own flaws.
One side is using data from income tax returns and the other is using data from census forms.
The reason that Burtless says that it is misguided or unfair for Reynolds/Rose to use census data to critique P&S's study is because the census data is not accurate for the top 2 to 2.5%, which is part of the main focus of the study.

On a sidenote... talk about good timing, I just got a present in the mail yesterday, and it brings this point home.
I've been selected for the second round of census data, the American Community Survey and am in the process of filling it out right now.
This thing is an f'ing book.
Am I seriously going to treat this the same as my taxes and go pull out all my bills and records for the last 12 months and take days to fill this thing out accurately, or am I just going to fill in a half-hearted guess and do this as quickly as possible and be done with it?
There's no penalty for filling it out wrong, like being audited, paying massive amounts of money or going to jail.

Given that fact, and your concern with garbage in/garbage out data, which data do you think is going to be more trustworthy or accurate? Census or tax returns?

He also confirms that the individual filer problem is part of his data:

"Because our data are based on tax returns, they do not provide information on the distribution of individual incomes within a tax unit. As a result, all our series are for tax units and not individuals."

I think we are arguing more than one issue here and they are becoming confused.
On one hand we are debating income inequality between the top 10% and bottom 90%.
No one is disputing that there is a growing inequality between the classes, just by how much.

On the other hand we were debating/clarifying your "majority of the lower 20% move to the top 20%" statement. (which I can't seem to find the basis for, other than that it is a repeated talking point by pundits)
Even so, I'm not going to dispute it.
I am willing to take your word for it because my point was when you look at the actual dollar amounts and income thresholds for it to be true, it is not that big a difference.
Now if you said that a majority of the bottom 20% moves into the top 10% I'd call BS.
Instead, I'm just saying that it doesn't mean what you seem to think it means.

No mention at all about his methodology for computing inflation, whether he used an updated pricing index, etc.

I've long since closed out all the pdfs but I do recall seeing an updated table that mentioned consumer price index which I believe may be the one that Rose was referring to that was changed to 50% the caveat being that it stated 50% growth for the top 1% instead of the top 10% as Rose's article implied.
Unfortunately I'm dealing with a death in the family so I don't have the time to go hunt it down at the moment, but I'll hunt it down when I get the chance.

But getting back to logic for a moment, if the whole study is done using the same parameters, is it going to change the results whether they equate everything to 2008 dollars or 2010 dollars?
Isn't inflation and consumer price index going to have the same affect on the value of $1 in the pocket of everyone equally?

Yes. But, I don't expect anyone to agree with me just because I say it. I expect them to agree because they also understand the evidence and logic for themselves.

For example, it doesn't require you to trust my opinion in order to understand how excluding massive amounts of income that disproportionately go to lower income groups would substantially alter the data.

Speaking of logic, in the scope of income data that we are talking about, how much of a difference are entitlements going to make between the lower 90% and the top 10%?

Yes, as a whole, the amount going to the lower class in welfare, unemployment, food stamps, housing assistance, etc. is a massive amount of money, but when broken down by household it is a very small amount.
No one is getting rich or moving to a higher bracket based on welfare checks and entitlements.
On top of that, most entitlement income counts as taxable income that gets reported so wouldn't it have automatically been counted in P&S's tax return data?

Even if it hadn't, does it really make logical sense that entitlements going to those near the poverty line are going to make a huge difference when comparing their income to multimillionaires?