Justus For All

None Sine Causa

Timely example

11:37 am on Thursday, November 30, 2006

This New York Times op-ed is exactly what I was talking about in this post from earlier today.

Despite significant gains in 2004, the total income Americans reported to the tax collector that year, adjusted for inflation, was still below its peak in 2000, new government data shows.

Reported income totaled $7.044 trillion in 2004, the latest year for which data is available, down from more than $7.143 trillion in 2000, new Internal Revenue Service data shows.

It has all the features that Alan Reynolds conclusively demolished as being an accurate way to analyze income data.  I have only read through part of the book once, and it is still on my nightstand so I can’t refer to it now, but it goes through several techniques that Reynolds completely destroyed.

I wish I had gained the expertise to demonstrate this myself, right now I only know enough to recognize this as bs when I see it.

5 Comments »

Comment by Alan Reynolds

December 4, 2006 @ 7:24 am

The New York Times piece about income reported on tax return simply proved what we already know — namely, that stock prices were higher in 2000 than in 2004.

Reported income includes one-time windfalls from exercised stock options and realized capital gains, both of which peaked with tech stocks in early 2000. Even as stocks started to fall in 2000 and 2001, there were still many long-term capital gains “realized” (cashed-in)from older investments as stockholders rushed for the exit.

As one of those investors, I reported the most “income” ever to the IRS in 2000, even as the market value of my lifetime savings fell by nearly a third.

Comment by probligo

December 4, 2006 @ 5:40 pm

Between the two of you, perhaps you could explain for this bear of small brain just how the government should report comparative income levels between years? Let us say over a 20 year period?

How might that reporting be conveyed in a manner that reflects what I am able to buy with my money.

For example, comparing (say) 2006 with 1996 and seeing a 50% increase in absolute income means very little if the cost of living has increased by 100% over the same period.

It should not require anything more than a simple statement or three, not a book-long exposition that will take me a month of Sundays to read.

Comment by Dave Justus

December 5, 2006 @ 8:36 am

This isn’t really about government reporting, it is about newspaper reporting.

Part of it is explaining what the known causes are. For example, the article implies that wages are stagnating, but the reported income being behind the 2000 levels is, and Alan explained, purely related to stock prices.

Depending on what you want to know, the ‘bast’ data for that is different. For example, if you want to know how well off people are, if their standard of living is improving or not, it is better to look at consumption data than income data, as consumption data smooths out a number of distoritions from income data. And obviously, when you adjust for inflation you should use the best possible measure of that, which is not always done.

Comment by probligo

December 5, 2006 @ 9:30 pm

Hmm.

Y’see, there is a difference.

There are fairly strict controls in NZ on the payment of remuneration by way of stock or stock options. The “income” that the stock represents is valued at the time of issue from the market (not face value), taxed at that time as income (or fringe benefit - that merely determines who pays) and it then takes the form of an asset in the hands of the employee. Because there is no capital gains tax, there is no further debate about the status of the stock or whether market gains and losses are “income”.

There was an attempt, about twenty years back, to introduce “employee share arrangements”. They never caught on; in part due to resistance from existing stockholders who feared dilution of the value of their investment, in part due to union resistance to the idea, and in part due to the likely tax regimes that would be imposed on the issuing companies.

Given that, I can draw a parallel as well…

There can be (will always be) variations in annual income. If the top 5% by value income earners were to experience a 20% change (up or down) it is an impact of 1% on the national totals. Take a different slice - if the top 10% of earners were to lose an average of 10%, then the effect is still 1% of total national income.

I submit that if the issue and payment of employee stocks as part of annual remuneration packages is included in the income numbers you discuss then that is quite valid.

On the other hand, if the figures you quote include a measure of capital value change then the article is misleading as you have said.

Please note that it is the inclusion of capital value change - whether realised by sale or notional by valuation - that is misleading and incorrect, not the comparisons made by use of inflation indicators or (the given rationale) the value of the stock exchange in each of the two years being compared.

To make it clear - if I own a parcel of shares cost $10,000 and the value of the shares on the market increases to $15,000 I have not “earned income” of $5,000. Similarly if the shares lose value I have not “lost” that value. It is only when I sell those shares that their current worth is realised (that might make you cringe but it is the approach used by NZ IRD and our tax laws). That truth is one of the reasons why NZ has never looked seriously at capital gains tax. The present NZ tax laws also allow the IRD to judge whether my portfolio and trading patterns are such that I am a “trader” or “speculator” rather than “private investor”. If I am a trader or speculator then they will come looking for income tax on my realised income, and that income will be included in the national income totals.

Comment by Dave Justus

December 6, 2006 @ 8:09 am

Hmm.. I don’t see how you could ever tax stock options as income at the time of receipt.

Regardless, given that the U.S. tax system does not work like New Zealands and this is based upon U.S. tax data. The higher market value, and the large number of stock options issued during the internet bubble made a big difference there.

I would guess though that at a market peak, and shortly after it in New Zealand a lot of ‘extra’ income would be realized as many people decide that that is a good time to sell (obviously it is) which would mean I would expect similar types of numbers with 2000 being a very high year for income, perhaps even higher in comparison to averages than the U.S.

Also note I never claimed that we shouldn’t measure inflation, but that their are several different measures of inflation that say different things. Obviously, using the measure that is believed to be most accurate is superior, but often not done.

RSS feed for comments on this post. TrackBack URI

Leave a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>