Two weeks ago we penned "As US Debt Hits New Record, Fiscal 2012 Tax Revenues Are 10% Higher Than Debt Issuance [7]" which unfortunately was very wrong: we completely forgot that tax revenues in the US are a two way street particularly from January through the end of tax season on April 15, when income and employment tax withholdings are offset by tax refunds as consumers rightfully claim (and in the process pad TurboTax revenues simply for having under-exempted themselves) what was overcollected by the government. Unfortunately, it also means that we showed the US in a far better fiscal light than it is in reality, because contrary to our conclusion that tax revenues are higher than debt issuance in fiscal 2012 (starting October 1, 2011), the reality is not only a mirror image, but worse, with total debt issued now surpassing net revenues (withholdings net of refunds) by a whopping 15%! In other words, for $710.7 billion issued in debt YTD (debt has risen from $14.79 trillion to $15.5 trillion), net tax revenues have risen only by $607 billion. Which means that contrary to conventional wisdom that the US collects in taxes modestly more than it issues, at least through the peak of tax refund season that is certainly not the case. It also means that little by little that neo-Keynesian ideal (where we hope we jest but are no longer sure) of all deficits being funded purely by debt issuance, is slowly coming true.
The chart below shows next tax revenues vs debt issuance YTD - go ahead and check: subtract year to date tax refunds from tax withholdings as of March 1 (link [8]) and compare to debt issuance since October 1, 2011 (link [9]).
[10]
Oops.
But wait, it gets worse.
Remember all that talk of a US employment based 'renaissance'? By definition, that would mean that more tax revenues have to be collected YTD compared to 2011, during which period the unemployment rate was logically far higher - after all it is a declining continuum, or so the BLS would have you believe. Because more people employed, means more taxes collected. Logic 101. Well, wrong.
As the next chart shows, comparing net withholdings, or total taxes withheld net of tax refunds, 2012 is now trending below the same period in 2011, by about $10 billion!
[11]
As a reminder, here is how the unemployment situtation stood at January 2011, when according to the BLS 130.5 Million people worked [12], when net government tax revenues were on par or better than 2012...
[13]
And compare this to the latest employment "data", when once again according to the BLS, 132.4 Million people worked [12].
[14]
So let's get this straight: America has seen the number of people employed rise by 1.9 million people from January 2011 to January 2012, and its unemployment "decline" by 0.7% in the same period, which means more taxes paid and thus withheld, and yet the tax collections have dropped from a period when unemployment was 9.0%?
Right.
Naturally, skeptics will say that this is purely a function of frontloaded tax refunds. Which would be great, however it is not true.
Stone McCarthy [15]deconstructs that myth as follows:
Cumulative individual income tax refund issuance for 2012 passed issuance for the comparable period last year for the first time in the latest week. In the week ended last Friday, income tax refund issuance totaled $26.6 billion, compared to $20,9 billion last year. That brought cumulative issuance to $85.4 billion, up 5.7% compared to the comparable period last year.
Today's Chart of the Day shows cumulative refund issuance through February 17 and for the comparable period for 2012.
Refunds are lagging our forecast, however. We had projected cumulative refund issuance of $102.3 billion through February 17. We forecast total refund issuance to be up 2.6% compared to last year. However, we expected more refund issuance early in the refund seasons compared to last year, when some returns didn't get processed until after February 14.
Refund issuance this year has been delayed by a different set of problems. As we have noted in our prior updates, the IRS has made changes to its systems to prevent fraud, and that caused refunds to go out more slowly than expected in the first few weeks of the refund season. The IRS appears to be catching up, but hasn't said so explicitly. The "Where's My Refund?" page simply now says taxpayers who file electronically can expect their refunds within 10 to 21 days. It will take another week or two before we can start to make reasonable comparisons between refund issuance this year and last year.
Finally, even if in reality the government is somehow goosing (i.e., frontloading) refunds, all it means is that cash available to US consumers is higher than where it should be in reality, as all that has happened is that the variable responsible for 70% of the US economy has obtained more cash earlier than when it should have been disbursed. Which in turn means that the recent Personal Income and Spending data [16], which was so disappointing it caused Goldman to lower its Q1 GDP tracking forecast [17](and that is even before the gasoline price shock), is even worse when one factors out the time effect of refund collection - traditionally an economic boost as the cash is spent as quickly as it is received.
So.... what is this about the US economy improving again?