By now, I think we are all familiar with the (correct) arguent that there is
no social security crisis. However, over at talkingpointsmemo, I saw
this from the
Times:
"I have a daughter at home. Her name is Wynter," said Ms. Jaques, sitting a few feet from President Bush at the White House economic conference on Thursday. "I want to make sure that she has Social Security when she retires as well."
Mr. Bush chimed in a moment later. "One of my visions of personal savings accounts is that Sandy will be able to pass her account on to Wynter as part of Wynter's capacity to retire as well."
This brought to mind a few questions...
First, who names their kid after Winter and then just misspells it? (Ok, that's my own pet peeve)
More seriously, this brought two questions:
- Given the current floated idea of letting people divert $1000/yr to a private account, how much money does this really get anyone? Is it enough to live off? Is it more than one would get from SS currently?
- Also, how likely is it that the sum of money is large enough that "Wynter" will get a penny in inheritence? In other words, is the investment gain large enough for an average person to live off of and still have some left over to pass along? Or, as I suspected, is this simply another way to give a huge tax break to the wealthy?
So, here's what I did. I found both historical inflation and stock market return data for about 100 years. (FYI the mean inflation is 3.55% with a deviation of 5.1% and the mean market return is 10% with a deviation of 15%). I then created a hypothetical account of a 20 year old who puts in $1000 a year every year up to age 68 when s/he retires. Then, I ran a monte carlo simulation to determine the probable outcome. Here's what I found for these assumptions:
mean $311,783.97
median $239,202.91
std dev $303,503.26
Now, for at 65, a reasonable expected lifespan seems to be about 15 years. Assuming , for ease of computation, that this person pulls all the money out at age 69 and spends it equally for 15 years (of course, with continued investment, the following numbers will be slightly higher). This makes a median payment of $1328 per month. This is not bad, current social security payments for retirees averaged $927 per month in 2004. However, we must realize that this is the median outcome, half the time, the amount will be above half the time it will be below. Again, the simulation showed that in roughly 18% of the simulations, the monthly payment would be around $600 per month.
Conclusion: The average person's median outcome is likely to be somewhat better than the current system but, and this is a big but, this person has assumed all the risk and must take the path with most risk and volitility- the stock market.
For the sake of comparison, I created a simulation with a more stable investment profile (mean investment return of 8% and standard deviation of 5%). The assumption here is that the individual hedges his/her bets to get a more stable return. Results:
mean $167,620.93
median $161,603.27
std dev $49,102.56
Again, with the same assumptions as above, the median monthly payment is about $898.
With this, albeit rather crude data in hand, let's answer the questions:
- The amount is roughly comparable to current SS benefits. This is not surprising, why would the administration contrive something that would be worse? The main issue is that is transfers all of the risk from the government to the individual investor. That's a good way to take the security out of social security.
- In the simulations, I assumed that all the money would be spend in 15 years. Leaving aside the obvious questions (what if one lives longer? shorter?), the simulations indicate that if a person had no other source of income, all the money would be used in 15 years and little or no money would be passed on the a beneficiary. Simply put, this is a huge giveaway to those rich enough to have other much larger retirement funds.
In fact, suppose I follow the initial assumptions with the goal of passing all of the private account to my child who will hold it to age 68. Assuming (so I don't have to do any more simulations), that I get an 8% after inflation return for my 15 years in retirement and she gets the same for 25 years after my death, the median sum of $239K rises to be $5,196,548. Not bad for a $49K investment on my part.
Let's all repeat this together:
There is no social security crisis!
For the median worker in the US, this plan may give roughly the same benefit as SS but at substantially more risk.
and, just like every other Bush program, this is another HUGE giveaway to wealthy people!
Update [2004-12-19 10:52:34 by Scott Pauls]: In the comments there is some decent discussion of what exactly the Bush admin proposes to do with the portion of your SS payroll tax that doesn't go to private accounts. I made the case that the Bush admin hasn't really made this at all clear and several times have floated huge reductions in benefits. doverpa take s more, shall we say, charitable view that the Bush admin will give you back what the rest of what you oput in according to the rules that are currently in place. I argues below that, either way, this always screws the bottom quarter of society and under the less charitible plan basically scres everyone except those who have abosultely no need for SS. Today, Snow came out with the latest line. Via pandagon:
John Snow on Fox News Sunday: Privatization is an add-on to Social Security, because you still have a claim on some Social Security.
This is actually an argument? Particularly against an uncharacteristically targeted question from Chris Wallace?
Snow admits we're going to have to reduce Social Security benefits by at least 2/3 for partial privatization of Social Security. Social Security is an insurance program, so when you cut the guaranteed insurance amount by 2/3, and then someone loses money on investments (because, as we all know, the "rate of return of the market" is a lot like "average tax cut" in the annals of dumb statistics), they can still rely on 1/3 of the former amount they were insured, alongside the nothing they gained from the market.
Editorial comment aside, Snow claims that future generations will only get about a third of current benefits. In the exampels discussed below, given the average retiree's current benefit of $928/mo, this means a future payout of $309. Thus to even break even, the private account must make up the $619. As shown by the simulations this requires one to take on maximum risk for the possibility of payout. Again, the simulation shows that even with maximum risk there is roughly a 10% chance that an entire generation will be below this line. Moreover, if one takes more reasonable risk (I, for example, reran the analysis with an industry standard of around a 6% mean return with a 7% deviation) on ends up with roughly a 40-50% chance of meeting or exceeding the current level of benefits.