Milliman on Asset Smoothing

It is difficult enough to place a value on trust assets that a public plan holds what with all those ‘alternative investments’ building up within most systems.  However when the asset number you use to determine funded status and annual ‘required’ contributions is some average of prior guesses that lately has resulted in ‘actuarial’ values of assets being massively higher than current market values for practically every public pension plan we have turned an actuarial assumption into an actuarial deception.

Quoting from the Milliman report criticizing Detroit’s official numbers:

the funded status of each System decreases significantly when measured on a market value basis.  In other words, there is a big hole to be dug out of.  Assuming that actual asset returns are as likely to be above as below the assumed rate of return in future years, asset returns are not expected to help the Systems dig out of the holes.  Cash contributions will be required.

For DGRS the smoothed asset value is roughly 45% greater than the actual asset value and for PFRS the smoothed asset value is roughly 30% greater than the actual asset value.  It is common for systems to have a “corridor” that constrains the smoothed asset value to within some range of the actual asset value.  A 20% corridor is very common (corporate plans are subject to a 10% corridor).

The smoothing period used for each System is 7 years which is longer than the common 5-year period.  This means that 2008-09 asset losses will not be fully recognized until 2016.

In the June 30, 2011 General Retirement System report (page E-2) GRS noted:

Present assets were reported to be valued using a seven-year smoothing of the difference between expected and actual investment income.

The smoothing process GRS used resulted in $2,421,566,956 in assets turning into $3,080,295,734, creating an extra $658,728,778.

The Asset Valuation method used by Milliman in their June 30, 2012 valuation of the New Jersey Teachers’ Pension and Annuity Fund (page 62):

A five year average of market value with write-up was used.  This method takes into account appreciation (depreciation) in investments in order to smooth asset values by averaging the excess of the actual over the expected income, on a market value basis, over a five year period.  Cash flows are based on an accrual accounting approach.  This method is prescribed by statute.

The smoothing process Milliman used resulted in $26,037,983,392 in assets turning into $31,079,212,983 creating an extra $5,041,229,591.

Asset smoothing might have had some useful purpose in the 1990s when double-digit earnings growth developed lower contribution levels that conscientious actuaries back then worried wouldn’t be able to be raised after the inevitable market correction so they manipulated asset values lower to maintain level contributions.  The ploy, as used these days, has been nicely debunked by Eileen Norcross at the Mercatus Center to which I would only add that public plans have no need for this type of chicanery except to gull an already innumerate public.  Governments have the ultimate contribution smoothing method at their disposal: without any funding rules they can arbitrarily name their contribution amount and that number can even be $0 in the worst-funded plan.

Every honest actuary (not in the pay of a government plan) will tell you that asset smoothing has no place in public plan valuations.  Milliman will even tell you that since, despite their warning that a corridor should be applied as they did in their New Jersey Teachers valuation, when it came to giving an ‘unbiased’ assessment of Detroit’s funding situation they valued the assets at………market.

9 responses to this post.

  1. I would disagree that smoothing has no place in public plan valuations. Smoothing is not some form of “chicanery” to avoid paying for the cost of a plan. Smoothing is employed to mitigate the impact of short term fluctuations from long-term cost estimates. As noted, smoothing helped prevent plans from taking advantage of the short-term run-up in asset values during the late 1990’s. It is appropriate that smoothing works both ways and helps reduce funding panic when there is a market dip like 2008-2009. Pension plans are long term (e.g., over 30-year) investments and a 5-year smoothing of short-term fluctuations is appropriate to help add predictability to pension cost budgeting. It sounds like the only critique Milliman had of the asset smoothing method was the lack of a corridor (e.g., smoothed value within 10% or 20% of market value). I would agree with that critique.

    Reply

    • I would disagree with that assessment for the simple reason that predictability is not an issue for public plans that requires gimmickry. In the private sector asset smoothing is acceptable since it is a means of stabilizing contribution levels. Private companies can’t pass a law putting a cap or a ceiling on their contributions or ignore IRC Section 430 so they are allowed this ploy. Governments, however, can.

      Reply

      • Posted by Tough Love on August 11, 2013 at 1:03 pm

        Smoothing WOULD serve a beneficial purpose in Public Sector Plan “IF” Public Plans HAD to fund their Plans per the actuary’s recommendations and the valuation assumptions (particularly the liability discount rate) were appropriately conservative.

        Of course THAT will never happen due to the “Moral Hazard” (the trading of generous pensions and benefits for Union campaign contributions and election support) that sweeps up our elected officials resulting in the granting of pensions FAR in excess of what is necessary, reasonable, and affordable to Taxpayers.

        Reply

  2. Asset smoothing produces a number called the Actuarial Value of Assets (AVA) . The AVA is an engineered value that modifies the true value of the plan’s assets, known as the Market Value of Assets (MVA). Some asset smoothing formulas use a “corridor” to limit how far the technique can depart from MVA (typically it limits variation to no less than 80 percent and no more than 120 percent of market value).

    Reply

  3. Asset & Liability smoothing is primarily used for the purpose of contribution smoothing – and hiding the true funded status as an implicit acknowledgment of weak government structures. If I can’t resist giving temporary investment gains away as permanent benefit increases when overfunded, that’s a telltale of weak governance – or not understanding that a plan with a highly volatile investment strategy needs a funding cushion in good economic times. The unfortunate side effect is that asset and liability smoothing negates the original purpose of going into high-risk strategies – higher average time-weighted returns are not converted into lower funding costs (aka higher money-weighted returns) because the assets are not there when the plan became underfunded and when we might have double digit returns for a while. Instead, there is revere-dollar cost averaging. DB plans in the US (and elsewhere) are dying because of a fundamental misconception – You cannot reach for higher expected returns (via a highly volatile investment strategy) to lower your funding costs and enjoy a smooth contribution path (achieved through smoothing). Both at the same time is impossible.

    Reply

  4. Excellent post! We are linking to this particularly great content on our website.

    Keep up the good writing.

    Reply

  5. […] makes them independent and whenever the word ‘actuarial’ appears in a report (as in actuarial value of assets) it often means ‘phony’.  This did not bother me so much until I got to do some […]

    Reply

  6. […] what public plan actuaries are paid for when right means higher contributions. Then there is the smoothing canard that the panel completely ignores, quoting the $44 billion actuarial value of assets as real rather […]

    Reply

  7. […] what public plan actuaries are paid for when right means higher contributions. Then there is the smoothing canard that the panel completely ignores, quoting the $44 billion actuarial value of assets as real rather […]

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: