The defined-benefit (DB) pension plan is beginning to play abigger role in liquidity forecasting and planning at $67 billionPfizer. The financial crisis and slide in interest rates createdvolatility in the funded status of many DB plans. Pfizer's plan wasfully funded in 2006, but fell to 81% funded by 2011. While thecompany has made cash contributions to the plan to shore up theassets and while asset valuations have rebounded, these pluses havelargely been offset by a sharp rise in pension liabilitiesresulting from lower discount rates.

Pfizer recently decided to move from a DB plan to an enhancedsavings plan, which will be completed by 2018. Until then, itneeded a sophisticated, reliable model to support planning forpension liquidity and its earnings-per-share impact. The pensionworld is full of predictive models, but Pfizer wanted somethingmore integrated and customized, so it built its own bettermousetrap at a cost of over 500 man-hours.

|

Pfizer developed the pension liquidity model over several yearsand implemented it in the second quarter. The model does notproduce a single forecast, but instead simulates thousands ofequally likely future scenarios with confidence ranges around thescenarios, explains Amit Singh, senior director of capitalmarkets. One big challenge is forecasting interest rates.“Due to their mean-reverting and stochastic nature, interest ratesare difficult to simulate,” Singh notes. Pfizer chose theHull-White method to calibrate to current market expectations forinterest rates because it is forward-looking instead of historical,unlike most interest-rate models, he says.

|

Predictions on asset classes and interest rates had to beintegrated and internally consistent. Simulating future values forasset classes and correlating them is common practice, Singhacknowledges, but leaving interest rates out of the correlation mixcan result in “nonsensical future scenarios where, for example,interest rates decrease in a simulated year but fixed assetsperform badly at the same time, something that would never reallyhappen,” he says. “We decided that rather than parse through allthe simulated paths and weed out the unreasonable ones beforeusing the rest, which would make the model tedious and sloppy, wewould create a technique that did not generate any unrealisticpaths by making interest rates a part of the asset correlationsmatrix. This was a breakthrough that made our model efficient.”

|

The project team spent time with members ofthe controller's group to be sure they understood the intricaciesof GAAP accounting and built them into the model. While Pfizer'smodel is independent, treasury tested its output against those ofother models to see if there were any wild deviations (thereweren't). The model is built on an Excel spreadsheet loaded withRisk^TM, an Excel add-in for Monte Carlo simulations, anduses FINCAD^TM financial software to build its Hull-Whiteinterest-rate model and calibrate it to current marketexpectations.

|

While the model's main purpose is forecasting, it also improvesasset allocation, which should bolster plan performance in the longrun. It is not an asset optimization model, however.

|

“Prior to this project, we had to rely on actuaries, bankers andinvestment managers to give us their separate model outputs,” saysassistant tresurer Neal Masia. “This was an unwieldy hodgepodge ofdata that we had to consolidate in-house to manage moreeffectively. Now we can take preemptive actions, plan for futureliquidity needs or adjust our current asset allocation to enhancethe likelihood of plan success.”

|

See a slideshow of the 2012 Alexander Hamilton Award winnershere.

|

Read about the Google project that won the Silver in theLiquidity Management category here and the Toyota Financial Services project that won theBronze here.

|