* Doug Pensinger ([EMAIL PROTECTED]) wrote:
> 1. How far would the tweaks I mentioned (and whatever other practical
> steps can be taken) do towards actually fixing the system as it now
> exists?
See below.
> 2. What is the advantage of having the Feds involved _at all_ in
> individual retirement savings that are not SS?
I don't think there is any. I would minimize it if I could. But some
people think there should be forced savings (in excess of the insurance
against poverty aspect). In that case, a fully-funded, privatized system
is superior to a pay-as-you-go system with a trust fund of ambiguous
economic signifigance.
> 3. Why can't we take whatever excess money there is _now_ in the
> system (including money that the Fed owes SS) and invest it in some
> secure way in order to insure SS's solvency?
There IS no excess money. The present value of projected future SS
receipts together with the trust fund balance is LESS than the present
value of future scheduled SS benefits. Over the next 75 years, the
deficit is about $3 to $4 trillion. If we don't artificially cut off the
deficits after 75 years, the total deficit is over $7 trillion.
In order to answer your first question, I need to give you some
background on my ideas and assumptions, probably more than you expected.
First of all, I have a rough idea of a long-term goal for humanity:
everyone able to do (almost) anything they want to do and no one needs
to take anything from anyone else in order to do as they wish. One
example of a civilization like I have in mind is "The Culture" from the
series of books by Iain Banks.
Obviously, we are not there yet. In the meantime, we have to allocate
limited resources as efficiently as possible to grow our civilization
towards long-term goals. But I (and I think most people) don't want
old, disabled, or unfortunate people starving in the streets, so while
providing incentives for maximum efficiency, there also needs to be a
safety net. Incidentally, providing a safety net is not necessarily
an efficiency hit since a good safety net should encourage healthy
risk-taking (in the interest of progress) that otherwise would not have
occurred.
But an adequate safety net that does not provide disincentives to work
productively is not easy to create. Historically, it seems that you
usually end up with a horrendously complex system with lots of special
circumstances and exceptions that result in a number of disincentives
for people to work efficiently or invest productively. For example,
if we have someone who could be working productively but does not
because of the safety net, then this is a disincentive that lowers
efficiency. One way this happens in practice is when welfare or benefits
are contingent on something, such as other income or having a job. If
you lose your safety-net payments because you saved heavily yourself or
because you took a job, then you have a disincentive to save or work.
Since complex bureaucracy almost always results in wasted (i.e.,
unproductive) effort, disincentives, and unintended consequences, I
favor simplicity. The most basic system I can think of is to provide
everyone who meets simple criteria (age, disability) with a minimum
amount of support in some manner (cash, vouchers, direct food and
shelter: various ways are possible). The key is to choose the support
level -- it should be affordable (resources ARE limited), it should be
humane, but it should be far from luxurious so that incentive to work
(or save) is preserved for those who are able.
Now I'll finally get to Social Security. But first I want to highly
recommend the book _The Coming Generational Storm: What You Need to
Know about America's Economic Future_, by Laurence Kotlikoff and Scott
Burns. It is a popular account (it only has a couple very simple
formulas in it, like A+B= +D) that is praised by Economics Nobel
Laureates George Akerlof, James Buchanan, and Paul Samuelson. It
criticizes policy mistakes of Democrats ("Mind you, the Democrats are
well aware of Social Security's structural problems. They've thought
about them carefully for a very long time and have come up with the
perfect solution: do nothing.") and Republicans alike ("although the
payroll tax rate rose every couple of years between 1950 and 1990, the
biggest increases occurred under Republican presidents...Eisenhower
3%...Nixon 2.7%....Reagan 1.4%..."). The book covers the aging problem,
generational accounting, Medicare, and yes, Social Security.
Social Security has a lot of problems. It is horrendously
complex. Quoting _tCGS_:
"The Social Security Handbook, which is supposed to clarify Social
Security's 2528 separate rules for paying taxes and receiving
benefits, runs for hundreds of pages. But talk to the benefit experts
at Social Security's administrative offices in Baltimore, and they''
just shake their heads about the accuracy of that document. For
certain questions, there may be only one old hand still knocking
about Social Security's massive administration buliding who knows the
answer....
The net results is that we have a black box handing out Social
Security benefits with very limited human oversight. A few years
back, the box was discovered to be making a big error in calculating
survivor benefits. From one day to the next, the government realized
it had underpaid beneficiaries (many of whom had already died) over a
billion dollars..."
But the worst problem is the pay-as-you-go method of finance. This seems
to appeal to the American culture of buy today, pay tomorrow, but it
is even worse than that: it is actually buy today, pay never. Leave
the bill for someone else to pay. When SS taxes and benefit levels
are simultaneously increased, young people get to pay the taxes and
old people receive the benefits. Membership in the AARP stands at 35
million people 50 and older who vote in large numbers, so the political
pressure will always be there to raise benefits (and payroll taxes). The
politicians will be glad to do it, since they are not accountable and
will be long gone by the time the problem they created blows up. For
example, the Diamond-Orszag proposal would raise payroll rates from
12.4% to a projected 15% by 2075 and add a 3% tax on payroll earnings
over the cap -- this would result in SS receipts increasing from 4.9% of
GDP today to about 6.6% of GDP in 2080. Any financial policy that that
demands an increasing fraction of GDP is obviously unsustainable. Look
at the history of SS, combined with the Diamond-Orszag plan:
----------------
SS
Year Receipts
% of GDP
================
1940 0.5
1950 0.7
1960 2.0
1970 3.2
1980 4.1
1990 4.9
2000 4.9
2010 5.1
2020 5.4
2030 5.6
2040 5.8
2050 6.0
2060 6.2
2070 6.4
2080 6.6
----------------
Obviously, this is not a sustainable path. But as long as we are on
a pay-as-you-go system, the political pressure is there to keep the
progression going, passing the bill to each successive generation.
So, here are the criteria that are needed in a reformed SS system:
1. Must be affordable, simple, and sustainable: no passing the bill
to the next generation
2. Provides a humane floor -- don't want to have elderly, survivors,
and disabled starving in the street
3. Preserve incentives for individuals to save and the nation to
invest
4. Any savings should be invested -- we need to raise the national
savings rate in a way that produces more productive investment
So, there are the assumptions that I am making. My conclusion is that
the best system should be fully-funded, not pay-as-you-go, otherwise
you will certainly fail condition 1. Also, minimizing the amount of the
benefits will help to meet condition 1.
Considering condition 2., the current system seems to provide an
acceptable level, so I would propose fixing the current benefit level
and only indexing it to cost of living for all future retires with
no wage adjustment (currently SS is indexed to wages before age 62,
and inflation after retirement). Also, I would not calculate initial
benefits based on earnings, but rather based on the minimum wage
regardless of a person's earnings, so everyone would get the same
amount. Raising the minimum wage would also raise the floor, so raises
in excess of cost of living would occur in a considered way (can we
afford it?) rather than automatically.
Number 3. is partially met by having the benefit paid as a floor
that is the same for everyone. Most people will want to save so they
can have a more comfortable retirement. In addition, I would end the
payroll tax and replace it with a national sales tax. Yes, a sales tax
is regressive, but so is the payroll tax it would replace. But the
sales tax is a consumption tax, so it helps encourage savings, which
we desperately need to do, and it may also help create jobs since
eliminating the payroll tax makes labor less expensive. Also, the sales
tax could actually go down, percentage wise, as the decades pass and the
real GDP grows. (If the sales tax idea is too big a change for people,
the system would work with payroll taxes combined with government
contributions for low income earners)
Condition 4. is where privatization comes in -- individual but
not individually controlled accounts. National sales tax revenues
would be contributed by the government into accounts earmarked for
each citizen. The accounts would be invested in a very low risk,
very broad set of index funds and bonds. Incidentally, many people
know that stocks have averaged a higher return than bonds over
the years, and that a portfolio made of 100% stocks is much more
risky than a portfolio made of 100% bonds. But it is a much less
known fact that, if you measure risk as standard deviation, then
portfolios with about 15% stocks and 85% bonds had both LOWER risk
and higher returns than a 100% bond portfolio in virtually every
historical period. I can't find a good return/risk graph with US
stocks and Treasuries on the web (although here is a close substitute
http://www.efficientfrontier.com/ef/797/expected.htm ), but I highly
recommend William Bernstein's book _The Intelligent Asset Allocator_ to
read more about this subject.
I would choose an allocation with only slightly higher risk than 100%
bonds, but much higher return: about 25% stock and 75% bonds. Most of
the bond portion could be "invested" in Treasuries but the bonds should
be real market-trading bonds. The equity portion could be invested in
a low-cost, global index fund run by a government agency created for
the purpose, or contracted out to third parties. (I like Bernstein's
idea of hiring Gus Sauter to run the agency -- for a few basis points a
year Sauter, or someone like him, and a small staff could easily invest
the money without any conflicts of interest since the allocation would
be decided by the market capitalization of the companies in the global
index)
The national sales tax should be adjusted so that there is a very low
probability that the savings accounts will not be able to provide the
required floor level. Since each account is individually earmarked, the
government is forced to adjust for changing demographics, and legal
methods could be devised to make it nearly impossible for the government
to "borrow" from the accounts to fund current spending. In the unlikely
event that there is not enough in the accounts to provide the floor,
then the government would be required to make up the difference out of
general revenues (perhaps a separate account could be established for
this purpose, and also to pay for survivors and disabled). The changes
could be phased in a little at a time rather than all at once, or some
parts implemented and others not. By the way, the system I described
borrows heavily from Kotlikoff's idea of PSS accounts.
Since the system I described is designed to encourage people to save for
their own retirement, it may be desirable to provide further incentives
for people to save (outside of the "floor" accounts). I would like to
see either an expansion of the Roth IRA, or even better, eliminating
taxes on capital gains and dividends and replacing the revenues with a
higher sales tax (and also sending everyone a quarterly tax refund check
paid from sales tax revenues to make the system less regressive) but I
digress...that's enough for now!
--
Erik Reuter http://www.erikreuter.net/
_______________________________________________
http://www.mccmedia.com/mailman/listinfo/brin-l
