Attacks on Social Security, Medicare borrow a strategy from Lenin

For three decades, conservatives' proposals for dramatic changes to
the programs have reflected a divide-and-conquer strategy inspired by
the Leninist movement.

By Michael Hiltzik

L.A. TIMES/January 13, 2012, 9:04 p.m.

About the last thing you'd ever expect is for conservatives to draw
procedural lessons from the founder of the Soviet state. So it's
fascinating to ponder the persistence of an attack on Social Security
that was explicitly billed as a "Leninist" strategy three decades ago
by analysts at the Heritage Foundation and is still in use today.

This is the notion, which is part of pretty much every proposal today
to "fix" Social Security and Medicare, that benefits for the retired
and near-retired should be guaranteed, while those for everyone else
must be cut.

The usual rationale given for distinguishing among generations is that
it's unfair to renege on a promise people have counted on for their
entire working lives. But the real rationale is political. If you
understand that, you might see almost all current proposals aimed at
reducing the costs of Social Security and Medicare — whether they
involve cutting benefits for most people across the board, raising
eligibility ages, or means-testing the programs to cut or deny
benefits to wealthier retirees — in a new light.

Let's go back to the original strategy brief by Stuart Butler and
Peter Germanis. Their piece, "Achieving a 'Leninist' Strategy,"
appeared in the Cato Institute's Cato Journal for fall 1983. Anguished
over President Reagan's failure to exploit Social Security's 1982
fiscal crisis to privatize the program, they concluded that the reason
was the program's strong support among the powerful voting bloc of
seniors.

The answer, they concluded, was to "neutralize" elderly voters while
continuing to undermine confidence in Social Security among the young.
Their model was the Leninist movement's "success in isolating and
weakening its opponents."

Any plan to change Social Security, they wrote, "must therefore be
neutral or (better still) clearly advantageous to senior citizens ...
the most powerful element of the coalition that opposes structural
reform."

The young, by contrast, were not organized to support privatization,
and uninformed about its virtues. The task of filling the knowledge
gap, they argued, could best be performed by "the business community
and financial institutions in particular ... both through their
commercial advertising and through public relations."

Ever since then, proposals for dramatic changes in Social Security and
Medicare have reflected this divide-and-conquer strategy. Some have
scarcely any other practical rationale. Consider, for example, the
argument for means-testing Social Security. This often appears as the
question of why the system should be burdened by paying a monthly
benefit to Warren Buffett or Bill Gates (insert name of your favorite
billionaire here).

Yet to reduce Social Security's costs significantly, any means test
would have to reach far beyond billionaires. One reason is that there
simply aren't enough taxpayers in the Buffett/Gates class to make a
difference, especially when the maximum initial annual Social Security
benefit, whatever your wealth, will be $30,156 this year. Only about
8,200 of the 140 million personal income tax returns filed with the
IRS in 2009 reported adjusted gross income of $10 million or more.

Moreover, Social Security benefits are already sharply skewed toward
the working class and middle class: 76% of all benefits paid in 2009
went to recipients with less than $20,000 in non-Social Security
income, according to calculations by Dean Baker and Hye Jin Rho of the
nonprofit Center for Economic and Policy Research. Those reporting
$180,000 or more got 1% of the total. In other words, means testing
makes no sense in terms of Social Security's fiscal condition; its
only result would be to make the program less relevant to the lives of
middle-class Americans — and that's a political strategy.

The same goes for increasing the full retirement age for Social
Security (currently 67 for those born in 1960 or later) and the
eligibility age for Medicare (65). An analysis of this commonly
discussed nostrum for both programs was just released by the
bipartisan Congressional Budget Office.

The CBO found that gradually raising the full retirement age to 70 for
those born in 1973 and later would indeed cut Social Security outlays
— by 2060 they would be 13% lower than if the law remained unchanged.
But the burden would fall especially heavily on low-income seniors,
who typically have few alternative income sources, and those for whom
staying in the workforce isn't an option. The CBO says the change
would "lower average income and increase poverty rates" among the
elderly; does everyone understand the trade-off of a policy change so
casually bruited about?

As for raising the Medicare age, that looks like a classic case of
being penny wise and just plain foolish. Raising the age to 67 would
reduce government expenditures on Medicare by 5%, the CBO says. But
the agency acknowledges that it would do nothing to stem overall
healthcare costs; indeed, its own analysis and those of other experts
suggest those costs would rise overall. Those who lose Medicare access
(those ages 65 and 66) "would pay higher premiums for health
insurance, pay more out of pocket, or both." And some would have no
insurance.

As the CBO observes, those effects would be moderated by the 2010
healthcare reform act, which many of those advocating cutbacks in
Medicare also say they want to repeal. The Kaiser Family Foundation
has concluded that pushing disenfranchised Medicare enrollees into
private insurance, either through the reform act's insurance exchanges
or by forcing them to stay on their employers' plans as workers or
retirees, would push up the costs of those plans by increasing the
numbers of elderly and less-healthy members in the private market.
Medicare premiums would also rise, because deferring the enrollment of
relatively younger beneficiaries makes the Medicare pool older and
sicker on average.

If the change were implemented all at once in 2014, the Kaiser study
found, the government would enjoy a net savings of $5.7 billion that
year — but at a total cost of $11.4 billion divided among 65- and
66-year-olds ($3.7 billion), employers ($4.5 billion), other Medicare
and private insurance members ($2.5 billion) and states paying their
Medicaid share ($700 million).

In other words, here as in so many other categories the savings from
making a dramatic change in an established program are illusory. Some
bargain. Lenin [sic] would be pleased.

Michael Hiltzik's column appears Sundays and Wednesdays. His latest
book is "The New Deal: A Modern History." Reach him at
[email protected], read past columns at latimes.com/hiltzik, check
out facebook.com/hiltzik and follow @latimeshiltzik on Twitter.

Copyright © 2012, Los Angeles Times

-- 
Jim Devine / "As far as the laws of mathematics refer to reality, they
are not certain; and as far as they are certain, they do not refer to
reality." -- Albert Einstein
_______________________________________________
pen-l mailing list
[email protected]
https://lists.csuchico.edu/mailman/listinfo/pen-l

Reply via email to