No Runny Eggs

Last week, I took a look at the ble…er, lack of future of the Disability Insurance (DI) portion of Social Security. Depending on the assumption of outgo used, that fund will likely run out of money between June 2015 and September 2016.

Since the Social Security Trustees are late with their already-delayed annual report, I decided to take a look at the much-larger Old-Age and Survivors Insurance (OASI) portion of Social Security. The Cliff’s Notes version, for those of you who don’t want to wade through the details on a holiday weekend, is that once again, the Trustees were overly optimistic last year. The OASI “Trust” Fund will begin running annual primary (cash) deficits this month, will likely begin running overall (including interest) deficits sometime between 2014 and 2020, and will likely be exhausted sometime between 2026 and 2033, with the latter two breakpoints depending on which rate-of-cost-growth one selects.

Like the look at the DI Fund, I assumed that tax revenues and outgo would do for the rest of 2010 what they did for the first 5 months (tax revenues decreased by 4.1%, outgo increased by 4.0%), that the interest rate would be 5% (a bit higher than the current weighted average of 4.880% and significantly higher than the 2.875% new bonds/certificates of indebtedness earn), and that tax revenues would increase by 5.3% starting in 2011 (ahead of the 1996-2006 average of 5.234% for the total FICA/SECA tax). Outside of a couple of “recovery” years, that 5.3% increase in tax revenues is also higher than anticipated in the 2009 Annual Report.

Because there will be a “bubble” of retirees over the next 20 years, I could not simply use either the rate of cost increase between the first 5 months of 2009 and the first 5 months of 2010 (like I did for the “base” DI scenario) or the average rate of cost increase between 1996 and 2006 (like I did for revenues). Instead, I calculated the annual increase of outgo using the percentages from the 2009 Annual Report, using a 2-year average for 2019 and 2020 and a 5-year average after 2020 because annual numbers were not available. Since the “low-cost” estimate is unrealistically optimistic (it claimed that even the DI “Trust” Fund would never run out of money), I used the “intermediate” and “high-cost” estimates for that, and ran the numbers twice.

As a review, the 2009 Annual Report claimed that, under the “intermediate” scenario, the OASI “Trust” Fund would run a yearly primary surplus until 2020 and a yearly overall surplus until 2024. Also, it would not be exhausted until 2041. Under the “high-cost” scenario, the yearly primary surpluses would last until 2014, yearly overall surpluses would last until sometime after 2018 (a specific year was not given), and the fund itself would last until 2031.

Fast-forward to this year. Unless something drastically changes between May and July, the 12-month primary deficit in the OASI fund will begin in the August 2009-July 2010 period (that’s this month, and 4 years before the high-cost estimate from last year), with the FY2010 primary deficit of $9.781 billion and a calendar-year 2010 primary deficit of $20.779 billion.

Assuming future cost growth at the “intermediate” rates, the yearly primary balance never quite gets out of the red in the intermediate term, coming no closer than $10.592 billion between March 2011 and April 2012. The yearly overall balance would flip to the negative side in 2020, and the fund itself would run out in 2033.

Before those of you who think that there really is a pot of gold at the end of every rainbow and that unicorn farts smell like Skittles start hammering me for being pessimistic, remember that I used a high estimate of tax revenue increase. Proof of that is the results of taking the estimates further out to 2085, which is as far as the Trustees go. Because, unlike the “intermediate” model from the Trustees, revenues increase faster than costs, this model projects the OASI “Trust” Fund goes back into the black in 2058. That is rather unlikely because under the “intermediate” model, the closest the OASI operation comes to running a yearly primary surplus after fund exhaustion is sometime between 2050 (when the primary balance is -2.98% of taxable payroll) and 2060 (balance of -3.06% of payroll), and likely near 2055 (balance of -2.96% of payroll).

Now, let’s take what the Trustees assumed to be the “intermediate” case for tax revenue increases from 2011 onwards and plug that back into the spreadsheet. Since they assumed roughly a 6% increase in 2012 and 2013, the OASI “Trust” Fund does get closer to the black on a yearly cash basis, getting as close as a $7.720 billion primary deficit between February 2012 and January 2013. However, they also assume a lower increase in tax revenues the other years, which means the fund goes into a yearly overall deficit in 2019, and runs out of money in 2030, never to go into the black again for more than one month in twelve (specifically, April).

It gets worse if future cost growth is at the “high-cost” rate. That is more likely given the first 5 months of 2010 have been more costly than even the 2009 high-cost estimate projected. Assuming either the flat 5.3% tax-revenue growth the “base” scenario does or the variable tax-revenue growth assumed by the Trustees in the 2009 “high-cost” estimate, the yearly overall balance would flip to the negative side in 2016, the last monthly overall positive balances in the intermediate term (long-term using the Trustees’ tax-revenue numbers) would be in April 2022 (on the strength of taxes) and June 2022 (on the strength of interest), and the fund would be exhausted in 2026.

Revisions/extensions (7:25 pm 7/3/2010) – The power of a HotAir-lanche is truly awesome. At last check, somewhere over 2,000 of you popped in here off the Headline Ed/Allahpundit put up. I can’t thank you enough for dropping in this little corner of the ‘Tubes, and Shoebox and I hope you stick around a while.

I know; the Trustees report is due out in a few days, but there’s some things that won’t wait until then. This time, I’ll focus specifically on the Disability Insurance (DI) portion of the matter. If you don’t want to wade through the whole post, the Cliff’s Notes version is that the DI “Trust” Fund suffered its first negative April since the effects of the previous major Social Security overhaul took full effect in 1987, and that it appears even the high-cost estimate in last year’s Trustees report was optimistic, with the DI fund now looking to be exhausted by the end of 2015 instead of by the end of 2016 in the 2009 high-cost estimate and by the end of 2020 in the intermediate estimate.

Before I get to the heart of the matter, I first need to do two months’ worth of housekeeping. In April, when just about everybody with a tax bill settles up and those who make quarterly estimated payments do so in the greatest of numbers, the combined Old-Age and Survivors/Disability Insurance (OASDI) funds took in $76,672 million, including $138 million of interest, and had expenses of $58,948 million for a gross increase of $17,724 million (or 23.12% of gross income), and a primary (cash) increase of $17,586 million (or 22.98% of ex-interest/net receipts). That pushed the 12-month primary deficit to $18,732 million, and dropped the 12-month gross surplus to $99,599 million (the first sub-$100,000 million 12-month gross surplus since 10/1997-9/1998).

In absolute terms, it’s the worst primary April performance since 1999 (with a $17,506 million increase on $49,810 million of tax revenue) and the worst gross April performance since 1998 (with a $15,502 million increase on $46,804 million in gross revenue). In percentage terms, that’s the worst April since monthly records were made available in 1987.

May didn’t show any real improvement over recent months, with the combined funds posting a combined $5,087 million gross/$5,177 primary loss on $53,859 million in gross receipts ($90 million in interest) and $58,946 million in outgo. The 12-month primary deficit grew to $21,987 million, and the 12-month gross surplus shrunk to $96,268 million.

This is not the first time since 1987 that the DI Fund faced imminent exhaustion. Indeed, in 1994, it came within 2 years of running out of Treasury securities to draw upon. The fix back then was to transfer the majority of the November 1994 tax revenue from the Old-Age and Survivors Insurance (OASI) Fund and increase the portion of the FICA/SECA tax that went into the DI Fund. That fix, which worked back then because the OASI Fund was running in the black, won’t work this time because, outside the “taxation” months of January and April, the OASI Fund has been running a primary deficit since July 2009.

Specifically to April, even in April 1994, the DI Fund ran a monthly surplus of $492 million ($473 million net) on $3,722 million of receipts ($19 million of interest) and $3,230 million of outgo. Now, the DI Fund has run 13 months of consecutive monthly primary deficits (through May), with the only monthly overall surpluses coming as a result of the semi-annual crediting of interest in June and December. The latest 12-month overall deficit (6/2009-5/2010) was $17,167 million, over 8.3% of the end-of-May fund value of $194,355 million.

That leads me to the bad news on the end of the DI “Trust” Fund. I decided to plug some numbers into the spreadsheet to see just how fast the fund would cease to exist. I assumed that tax revenues would, for the rest of 2010, continue to lag behind 2009’s by just under 5.2% (as they have for the first 5 months), then increase by 5.3% (actually a bit higher than the 1996-2006 annual average of 5.234% for the total FICA/SECA tax). I also assumed that outgo would increase by the 5.464% that it went up for the first 5 months in 2010 versus the first 5 months of 2009 (significantly lower than the 1996-2006 annual average of 7.613%), even though 2010’s increase is solely due to the increased number of those drawing from Social Security, and that the interest rate would be at 5% (a bit higher than the current weighted average of 4.880% and significantly higher than the 2.875% new bonds/certificates of indebtedness earn).

With a starting point of $194,355 million in the DI “Trust” Fund at the end of May 2010, and assumptions that are actually favorable to the longevity of the “Trust” Fund, here are the projected DI “Trust” Fund balances at the end of each calendar year:

December 2010 – $178,728 million (projected to be the first “interest” month with an monthly overall deficit, with June 2010 having the only monthly overall surplus of 2010)
December 2011 – $151,653 million (again with June 2011 having the only monthly overall surplus of 2011)
December 2012 – $118,899 million (with June 2012 having the only monthly overall surplus of 2012, and June 2012 also having the last monthly overall surplus in this scenario)
December 2013 – $83,281 million
December 2014 – $43,428 million
December 2015 – EXHAUSTED (November 2015 would see a DI fund balance of $4,839 million)

Even if I were to assume the 5.3% tax growth beginning in June 2010 instead of January 2011, it doesn’t get much better. The DI “Trust” Fund barely makes it into 2017 before it disappears.

As a reminder, under current law, at the point the DI fund is exhausted, benefits would have to be cut by the percentage the tax revenues don’t meet demand. No, it can’t even tap into the OASI “Trust” Fund to avoid average benefit cuts of over 25%.

Revisions/extensions (7:05 pm 7/2/2010) – Partly because the Social Security Administration missed their delayed deadline for the annual report, and partly because I had to adjust my outgo assumptions for the OASI “Trust” Fund estimates, I have re-run the numbers using the increases anticipated in the “Intermediate” and “High-Cost” scenarios in the 2009 Trustees Report.

Since the Intermediate scenario assumes that the cost of running the DI portion of Social Security rises slower than the 5.464% it went up in the first 5 months of 2010, the DI “Trust” Fund makes it to December 2016 before being exhausted. Further, unlike the “base” scenario outlined above, the DI operations would be able to pay out the entirety of the scheduled payments for a couple Aprils after fund exhaustion.

The bad news is the “High-Cost” scenario assumes the cost of running the DI portion of Social Security rises even faster than my assumption. Plugging that back into the spreadsheet yields fund exhaustion in July 2015.