Send me real-time posts from this site at my email
Alex Cho

Why Savers Are Losers And What Corporate America Is Doing to Your Pension

Going into Q1'16 earnings season I wasn't anticipating there to be additional earnings catalysts, but according to Credit Suisse, pension cost recognition could meaningfully swing the S&P 500 earnings figure for the duration of 2016. This could be a needle-mover, and to fully understand what I mean, I need to explain some actuarial science to you all, because the impact on sales and earnings is worth thinking of in the context of S&P 500 aggregated earnings.

Generally speaking, the issue pertaining to pensions is whether or not they're underfunded or overfunded liabilities. In other words, there are assumptions attached to a long-term liability, which is a pre-defined benefit an employer must pay their employee upon retirement. Assuming the employer can earn a higher interest rate on the assets in the plan, the company can quickly fulfill pension liability obligations.

However, things have gotten a little tricky with the accounting of pension expenses.

Pension accounting utilizes accrual accounting to recognize an expense that has yet to occur until the payout period. Nonetheless GAAP will immediately recognize the retirement expense in the current accounting period even if the benefit payouts were to happen many years down the road.

In the past six-months, the FASB (financial accounting standard board) has shifted its stance on interest cost and service cost. I quote PricewaterhouseCoopers on the specifics of the subject matter:

In developing the Interest Cost and Service Cost components of net benefit cost for a defined benefit retirement plan under US GAAP, one key assumption is the discount rate. We understand that many companies and their actuaries have been considering various alternative approaches to determining the discount rate for purposes of calculating Interest Cost and Service Cost. In evaluating a specific fact pattern, the SEC Staff recently indicated it would not object to a registrant changing to certain alternative approaches, including the use of individual spot rates (referred to below as the spot rate approach).

So what the hell does all of that mean? In a nutshell, S&P 500 companies can alter their discount rate lower to reflect benchmark interest rates, i.e. shorter duration treasury securities. The interest cost is the minimum return required to meet long-term pension obligations, but the input determines the output. So if you input lower interest cost what you get out is a lower benefit obligation. I.e. the employers are now able to offer smaller pension benefit payouts in response to a weak yield environment, and if the pension assets are no longer sufficient to cover long-term liabilities, the employer will then renegotiate with the labor union to arrive at a healthier funding rate to pivot from underfunded to overfunded. 

Source: Investopedia

How does any of this matter? Well, according to PwC the discount rate assumption can be tied to alternative spot rates, which lowers the assumed interest cost. The lower interest cost is essentially what actuaries forecast years out into the future based on current benchmark rates and then discount back to present value to determine the current cost of pension interest liabilities. The service cost reflects the incremental increase to net pension obligations for every additional year of employment, so you're adding in long-term pension obligations and discounting back to present value the pre-existing obligations that have yet to be paid out concurrently when determining projected benefit obligations. The actuarial gain/loss reflects the anticipated returns in excess of interest cost, which is often used by employers to deflate the required obligations in the present accounting period. So if your benchmark is 3.4%, but your expected return on pension assets is 7% to 8% you're able to reduce pension benefit obligations by setting a low bar for required return, but a high bar on what you're actually going to return. Now, these actuarial assumptions are pretty aggressive and earning yields in excess of 5% isn't attainable on an investment grade basis. But as you will notice in the disclosure of Alcoa, they've come away with a really wonky way of proving that they can earn high returns on investment grade bonds. 

The present value of a pension asset reflects the return on assets, contributions minus the benefits paid. This is straight forward, and isn't really debatable. Much of the accounting/actuarial trickery occurs on the right side of the diagram (when determining pension benefit obligations).

So, here's what Alcoa states as their interest cost:

In 2014, 2013, and 2012, the discount rate used to determine benefit obligations for U.S. pension and other postretirement benefit plans was 4.00%, 4.80%, and 4.15%, respectively. The impact on the liabilities of a change in the discount rate of 1/4 of 1% would be approximately $460 and either a charge or credit of approximately $20 to after-tax earnings in the following year.

Here's what Alcoa projects for returns on plan assets:

For 2014, 2013, and 2012, management used 8.00%, 8.50%, and 8.50% as its expected long-term rate of return, which was based on the prevailing and planned strategic asset allocations, as well as estimates of future returns by asset class. These rates fell within the respective range of the 20-year moving average of actual performance and the expected future return developed by asset class.

So, when we subtract 8% from 4% we walk away with the assumption that Aloca can fund half of its obligations due to net actuarial gain! That means the actuaries honestly believe they can outperform the current yield environment by a full 4% by averaging bond yields over a 20-year rolling period! What absurdity! If anything, we will be lucky to see interest rates on investment grade corporate bonds average a 200 basis point spread above the 10-year treasury assuming the 10-year treasuries ever reaches 5%! But wait…. It even gets better, because the SEC is now onboard with this practice too, so this is a big screw you to working class America as companies can reduce interest expense assumptions and inflate return assumptions to reduce their pension benefit obligations! Assuming they're wrong, which they will be, they will then tell labor unions that they need to renegotiate pension benefits even lower.

So, tell your friends if you truly care about them, that pensions won't guarantee you a good retirement.

Here's what Credit Suisse anticipates for the discount rate:

As you can see in Exhibit 6 by having the weighting shift from the total benefit payments (which peak around 2035) under the traditional approach to the present value of those payments (which peak now) under the spot rate approach, you end up with a lower interest cost related discount rate (shifting from the longer end of the yield curve where rates are higher to the shorter end where rates are lower). As a result, the discount rate fell from 4.33% to 3.62% in this example.

In the earlier example of Alcoa, the discount rate was 4%, and with the ability to use a weighted average of spot rates, the same pension liability is discounted to present value with 40 basis points lopped off. When you have a cumulative 40 basis point reduction over a 15-year period, the net impact on obligations is a 6.71% reduction, because when you compound 40 basis points over a 15-year period, the expected payout is an additional 6.17%, which is something Alcoa won't honor as a total percentage of obligations over a fifteen-year period because they expect a lower interest expense. They will then use an expected return of 7% to 8% over the next 15-years to rationalize why their pension obligations are funded, which is another way of saying that the company will set aside less cash for employer contributions thus inflating earnings for 2016. Chances are they will renegotiate based on their actual pension assets at some future point, which goes back to why pension-based savers are losers in America.

Credit Suisse mentions that this will positively impact S&P 500 earnings over 2015:

For example, pretax income for the S&P 500 would have been 1.7% higher, which doesn't sound like much but it could be significant depending upon your expectations for earnings growth in 2016. When it comes to individual companies, our back of the envelope estimates suggest that there are 48 companies where pretax income would have been more than 5% higher by switching to the spot rate method.

Assuming S&P 500 (SPY) earnings net a 1.7% positive impact on EPS growth, and when combining Factset figures for CY 2016, we arrive at an EPS growth rate of 6.71% + 1.7%, which totals to 8.48% earnings growth for the S&P 500. This translates into $127.70 EPS for CY 2016 for the entire S&P 500 component as opposed to $125.71 EPS for the current Factset estimate. Yes, I'm well aware Credit Suisse is referring to pre-tax, but the impact isn't that noticeable, and for sake of simplification I'm directly inputting the 1.7% positive EBT assumption.

I don't anticipate the analyst consensus to factor the full impact of pension obligation reductions into current year figures, because it's not a needle mover in the vast majority of instances as many opt for linear extrapolations on cost assumptions when pertaining to SG&A. Therefore, it's a sleeper catalyst, which will achieve full implementation gradually among the S&P 500 companies over the next four quarters, which may result into reclassification on prior 10-Qs.

Therefore, I'm going to forecast S&P 500 earnings using a 17.4 multiple to arrive at a valuation of 2,221.98 for the S&P 500. This would compare to 2,187.44 without the recent accounting changes to pension benefit obligations.

The market could return 16.2% this year assuming mean reversion, and a consistent pattern of better than expected earnings. I'm fairly certain the market will struggle in the earlier parts of the year, but a path to recovery will emerge for names with better fundamentals with energy the primary laggard for 2016.

If you're really serious about retirement, I think it's time to consider a 401(k) over a company sponsored pension plan because the money public companies save on pensions will directly correspond to stock returns in 2016. Therefore, investors would be better positioned in stock based mutual funds as opposed to bond based pension funds, because companies are cutting a lot of corners to keep pension costs down, and if they remain underfunded further on down the road they will renegotiate lower benefit payouts with labor unions on aggregate.

Welcome! Is it your First time here?

What are you looking for? Select your points of interest to improve your first-time experience:

Apply & Continue