Social Security

The October 2013 Social Security recipients report was released today showing a monthly increase of 110k people with an average monthly benefit of $1,163…up $1 over September. We continue to see a gradual slowdown…with 20k fewer new enrollments than last October’s 130k increase. The TTM delta is now at 1.227M, so averaging right at 100k per month, though there are large seasonal variations (which is why we look at the TTM).

For reference, we peaked at a TTM rate of 1.624M almost 4 years ago in 12/2009 and have been trending down since then. Still…1.227 is still very high historically. Looking back 10 years to 2013, the rate was under 600k. The course from here will be a very important factor in how quickly the budget deficit deteriorates over the next 5-10 years. On the positive side, though high, the rate is still trending down, and there is a good chance that a lot of the people you would have expected to be retiring now or soon actually already did….causing the 2009 spike…and the subsequent dip.

On the other side, we are still working through the left tail of the boomer  population distribution. I have a table published in 2011 that shows the US population by age. As of 2011, there were 2.7M of living Americans born in 1948….who would be turning 65 this year. Contrast that to those born in 1949 at 3.7M. Not sure what they put in the water that year, but roughly 1M more Americans will turn 65 in 2014 than in 2013. From here, we have a steady increase until 2028 when 4.6M born in 1963 will turn 65. Obviously that’s based on the 2011 snapshot and not adjusted for mortality. And while currently, people become eligible for Medicaid at 65, they can apply for SS anytime between 62 and 70…with the benefits growing the longer they wait.

So the main thing I am looking for with this data series is a bottom in the TTM rate and for the trend to turn back up. October 2013 was not that month. I don’t know if it happens next year, or five years from now, but when it does lookout because it’s going to get ugly(er).

The August Social Security monthly update came out last week, but I am just now getting a chance to look at the latest results. If you aren’t familiar with the report, basically, it gives us the number of participants in the Social Security program, both the Old Age and Disability portions of the program. It also gives us average monthly payouts per category, and some demographic male/female breakout.

So…in a given month, the given population changes in two ways. First, new people reach retirement age or become disabled, adding to the population. Second, retired workers pass away, or disabled workers recover and join the workforce. Just to frame it, the current population on Social Security is 57.554M or 18% of the total US population and the average monthly payout is $1,162 per month.

In my prior two posts on June and July, I focused primarily on the retired worker population…by far the largest at ~37M. To continue that analysis for August, the monthly change was an increase of 71k over the prior month ending up at 37.575M people compared to last months increase of 112k people. So great news right? Well…not exactly. See,  there is a lot of seasonality involved, driven by retirement and death patterns that cause a lot of variation in rates over the year. December is generally the lowest,  at 59k in 12/2012. January is typically the highest at 171k of additions in 1/2013. The rest of the months fall somewhere in between. So…rather than comparing the monthly change to the prior month…it is best to compare to the year ago delta.

August 2012 had an increase of 85k, so this is an actual improvement, bringing the TTM change down a hair from 1.191M to 1.177M, still within a stones throw of the December 2009 1.240M record. So despite the small improvement, we are still adding people to the program at nearly a record rate and this rate will almost certainly continue to trend up over the next decade as we get into the meat of the Boomer retirements.

This is the same chart we’ve seen in prior posts, but I’ve narrowed the range only showing 2005 to present. You can clearly see the 2005-2008 average of about 500k doubling during the great recession to the 12/2009 peak, trending down, and then back up to the current peak just under 1.2M per year rate. You can see the small dip at the end….I wouldn’t get excited about that yet…but we will keep an eye on it.

Now, lets take a step back and look at the entire population. It includes disability, currently at 8.9M and growing at about 12k per month over the last year, as well as various subcategories of spouses, widows, and children. The entire population grew 82k to 57.554M, down from the year ago increase of 103k. Lets take a look at the chart:

This chart looks a bit less menacing. After following the same trend up to 1.6M per year the rate is has steadily declined to around 1.26M per year. As it turns out…most of the other categories seem to have more or less stabilized, with retired workers, and to a lesser degree disabled workers being the only materially changing categories. Still… 3 years past the great recession, we are nowhere near to the normalized rates we saw leading up to 2008. Bottom line, the TTM cash cost of SS related programs was $710B over the last 12 months, and is growing at an annual rate of $60B. By the time the next president takes office in Jan. 2017, that annual rate will be around $922B. By Jan. 2021…it will be closer to $1.2T.

So…to wrap it all up, 85k added to the welfare rolls in August 2013. The overall rate slowly is declining…for now, but the annual add is still nearly twice the pre-recession rate, and likely to turn back up before too long as the Boomers keep retiring in droves. There is no happy ending here…it will just get worse and worse and worse until it is ultimately defaulted on with the rest of the debt.

The Official July Social Security update was released this morning so I thought I’d take a look and give an update…here is my June update if you missed it.

The chart is largely unchanged. For July, we added 112k new retired workers to the population…I don’t have the data to confirm this, but generally…you are looking at about 300k new enrollees, being offset by ~200k deaths. Also note that additions seem to be driven seasonally with Jan being the highest month at +171k this year, and December being the lowest…59k in 12/2012. I don’t know if this is driven by retirement patterns or death rate patterns, I suppose it doesn’t really matter.

What we are really looking at then…is to compare July 2013 to July 2012. It turns out they are nearly identical…July 2012 had 110k of additions, so less than a 2% increase. This very may well be an anomaly…the last 4 months have averaged almost 17% YOY gains. With minimal YOY change…our TTM…charted above rises the 2k…from an annual rate of +1.190M to +1.192M

The average benefit rose $0.86 from $1269.38 to 1268.52. This sounds inconsequential, but it suggests to me that new retirees are on average coming in with higher monthly benefits, as much as $300 higher than the beneficiaries that are passing away. This could be driven largely by narrowing of the gender gap…both in labor force participation and pay. So just imagine that in a given month, a 95 year old female beneficiary passes, and a 65 year old female  applies for benefits. With 30 years between them….they likely have very different work histories….and very different $ benefits. In this scenario, even though the count stays the same, the average could go up quite a bit. In the long run, we can likely expect the current $300 gender gap ($1425-male $1111-female) to close, and perhaps even reverse itself…and while this may be a good thing for society….it will add additional strain to social security. If the gap were eliminated today….monthly benefits would jump nearly$6B…$70B per year (or $700B over ten….which is how congress likes to digest it’s data)…all of course added directly to the deficit.

So bottom line, we are currently adding a net of nearly 1.2M people to the social security retirement rolls per year, a trend that should start ramping up over the next five years as we get into the meat of the boomers. This may be a slow motion trainwreck, but the end result is going to be just as ugly.