Would You Like To Take A Pension Survey?

I loved the Animaniacs cartoon when I was a teenager, especially one episode called “Survey Ladies.” In it, two ladies run around a shopping mall hounding the Animaniacs screaming, “would you like to take a survey?” and asking crazy questions like, “would you eat beans with George Wendt?” For those of you who don’t know, George Wendt was Norm in the sitcom Cheers.

That’s how I felt in March 2021 as I administered a pension survey to US-based pensionable workers and retirees from several personal finance Facebook groups and my blog’s email distribution list. I ran around (virtually) trying to convince pensionable workers and retirees to take my survey and answer many seemingly crazy pension-related questions. Sadly, I couldn’t figure out a way to work George Wendt or beans into it.

https://youtu.be/Bwi96ZRtc_8?t=73

Now, some of you may think that I love defined benefit pensions (DBPs) so much that I’d run a pension survey just for kicks on a Friday night. But, in reality, I only ran it to meet the ‘primary data collection’ requirements for my master’s thesis. That said, I could’ve chosen a different means of collecting data, like interviews. However, since I had access to numerous pensionable workers and retirees through my writing and social media presence, I decided a survey might provide some fascinating insight into my readership. In that regard, it didn’t disappoint.

Participants

Ultimately, 306 US-based pensionable workers and retirees participated in my survey in one form or another. Participation wasn’t random, nor was the population sample representative of pensionable employees and retirees everywhere. At best, the 306 who participated, and the 130 who qualified to answer all the questions, represent my average readership. In other words, people potentially like you.

On average, the 306 participants were between 35 and 54 years old, female, white, married, well educated, and pensionable employees. The table below provides the exact demographic breakdowns:

Survey Intent

I created and administered the pension survey to measure the impact of DBP plan design on pensionable employees’ voluntary turnover decisions. I expressly set out to determine which individual pension design elements made pensionable employees consider staying the most during their Golden Albatross (i.e., stay-or-go) decision. The survey achieved this by asking qualifying participants to rank up to three design elements that made them consider staying during their voluntary turnover decision. For the ranking portion, it didn’t matter if they ultimately stayed or not. I simply wanted to know which design elements made them think about staying the most.

It’s important to note that the pension design elements I studied were the same pension design elements I discussed in Part 25 of the Pension Series. Furthermore, as discussed in Part 26 of the Pension Series, voluntary employee turnover means an employee chooses to quit a job. It doesn’t include layoffs or retirement. If you haven’t read those posts yet, the results from this survey won’t make as much sense, so I suggest you read them now.

Explaining The Results

The survey produced a lot of data. So much data, in fact, that I split the results over several blog posts. As a result, in this post, I highlight what I consider the most important and exciting findings from the ranked preference portion of the survey. Part 28 of the Pension Series covers the results from my statistical analysis of the survey data.

Not to insult anyone’s intelligence, but I tried to break down the below highlights into layman’s terms to make them easier to understand. For those who want to read everything in mind-numbing, academic-style prose, I embedded a sanitized copy of my thesis below this paragraph. In it, I discuss such wonderful things as quantitative analysis and p-values. It’s a great read … if you’re suffering from insomnia.

Ranking Results

Right, let’s discuss the results from the ranking questions. After all, people love to rank stuff. They especially love to look at ranked results like elections and opinion polls. I’m no different! So, with little explanation, here’s a graph that answers which pension design elements made pensionable workers consider staying the most during their Golden Albatross moment. It reads from left to right, most influential to least influential:

Context

OK, now it’s time for the context. At this point in the survey, only 144 of 306 respondents remained, and only 130 qualified to rank design elements. First, I asked the 130 respondents to identify the pension design elements in their pension plans at the time of their Golden Albatross decision before ranking them. After all, I didn’t want anyone ranking design elements that were not in their pension plan. If participants couldn’t identify the design elements within their pension plan, they could skip this part of the survey, which is why 14 of the 144 did not rank design elements.

As you saw in the above graphic, when available, the pension subsidized healthcare, immediate annuity, and low-risk retirement income design elements (from Part 25 of the Pension Series) prompted pensionable employees to consider staying the most during their voluntary turnover (aka Golden Albatross) decision. The key phrase is “when available,” since, of course, not everyone had all (or any) of these design elements available in their pension plan. In fact, as the graphic below depicts, 67% of voters had healthcare available. In contrast, only 45% had immediate annuities available, and 60% had low-risk retirement income available. Thus, since every design element had a different number of voters, I had to determine how to compare the ranked results from each design element that didn’t simply involve weighting and tallying up the total votes.

Efficiency Scores

I solved the “different number of voters” problem in two ways. For the first solution, I averaged the number of weighted points per voter each design element scored. I called this voter efficiency (Ve). For the second solution, I took the total weighted score from each design element and divided it by its maximum theoretical score, much like a school teacher would do on a test. I called this element efficiency (Ee).

Both efficiency methods ensured I compared apples to apples, and both produced the same ranked order. The Combined Respondents’ Efficiency Score graphic (above) displays those results and provides the definitive rank order based on survey participants’ stated preferences. The below chart shows the scores from both methods for each design element. The chart also lists the design elements from highest efficiency scores to lowest.

Healthcare Analysis

Now that I explained how I got the results let’s talk about what those results may mean. Again, if you haven’t read Part 25 of the Pension Series, you need to because I’m about to refer to it heavily.

The healthcare result is not surprising for a survey population made up of US-based pensionable employees and retirees. As noted in Parts 6 and 25 of the Pension Series, healthcare is expensive in the US, and costs have exploded over recent decades (Liang et al., 2020). For this reason, in 2017, healthcare was ranked the number one overall preferred employment benefit in the US (Jones, 2017).

As my survey results show, pension subsidized healthcare is also a popular defined retirement benefit. This is because it stands to save pensionable retirees tens to hundreds of thousands of dollars as they age in retirement (Hoffman & Jackson, 2013). Moreover, healthcare costs are somewhat unpredictable. Granted, as discussed in Part 25, households spend increasingly more on healthcare as they age. But, from year to year, no one knows what ailments will require medical care, which makes budgeting for it hard.

That’s the main reason why healthcare coverage is pooled as insurance in the US. It spreads the risk of needing large amounts of money to pay for medical expenses among people in varying degrees of health. Doing this limits the budget impact of that unpredictability on any given individual. Thus, it’s no surprise that pensionable workers with access to a pension fund (or an associated healthcare trust fund) that provides subsidized healthcare coverage favored it so heavily. So much so, that healthcare’s element efficiency score was 10% higher than the second-place design element, immediate annuity.

Immediate Annuity

Speaking of the immediate annuity design element, it was the second most influential design element. By analyzing the demographic and ranking data, I ascertained that the survey participants in the US military’s High-36 (aka High 3) pension plan drove this result. I was able to do this because I asked ranking participants what type of pension plan they belonged to. 33 of the 59 participants with access to an immediate annuity came from the High-3 system.

As noted in Parts 3 and 25 of the Pension Series, immediate payout annuities are expensive for employers like the US military to provide (Enns et al., 1984). They are also lucrative for retirees because they pay out more money over longer periods than other forms of pension annuities (Benartzi et al., 2011). Believe it or not, most military ‘retirees’ understand this math intuitively, which also makes this ranking result unsurprising.

I say that the result is unsurprising because the vast majority of US military personnel who make it to pension eligibility ‘retire’ shortly after the 20-year full vesting mark (Enns et al., 1984; Sharp & Biderman, 1966; Warner, 2008). They often do this because they know how lucrative that getting paid by their previous employer to not work for the remainder of their lives can be. Moreover, most ‘retirees’ usually go on to second careers which allow them to collect two paychecks, one from their pension plan and one from their second career employer (Enns et al., 1984; Sharp & Biderman, 1966; Warner, 2008).

Low-Risk Retirement Income

Low-risk retirement income‘s third-place finish is unconventional since it is not an intentional design element but a by-product of good pension fund governance. The survey defined low-risk retirement income as “ultra-safe like US Federal pensions as well as corporate and public pension funds at or above a 100% funding ratio.” Now, theoretically, all DBP plans are supposed to provide risk-free or low-risk retirement income to their members (Jennings & Reichenstein, 2003). However, as discussed in Part 1 and 25 of the Pension Series, that is not the case in practice.

Many US pension plans’ funding ratios decreased severely over the past two decades (Aubry et al., 2018). Furthermore, pension scholars have noted a clear separation between the US’s worst and best-funded pension plans (Aubry et al., 2020; Aubry & Wandrei, 2020). This is no secret since US media headlines often contain stories about poorly funded pension plans, especially at state and local levels. As a result, it’s not surprising that the potential for low-risk retirement income drew participants who believed that their pension plans included this design element towards staying.

Continuance Commitment And Pension Quit Costs Review

OK, let’s break from examining each individual result and discuss the overall results instead. In Part 26 of the Pension Series, I explained the theory of continuance commitment. I also explained the idea behind pension quit costs. In fact, in that article, I specifically stated that I used the continuance commitment and pension quit cost theories as a framework for my thesis. Therefore, I built my survey in such a way that I could compare the ranking results against that framework.

As a quick recap for those that don’t remember or didn’t read Part 26, Meyer and Allen’s (1991) original codification of continuance commitment included an employee’s recognition of the costs associated with voluntary departure. If the employee determined the cost of leaving was too high and stayed with the organization, they became continually committed. Once those costs drop to an acceptable level or cease to exist, continually committed employees will leave. The specific mechanism that some researchers believe pensionable workers use to calculate those costs is pension quit costs (Ippolito, 1991, 2002; Nyce, 2007; Luchak et al., 2008). Using this mechanism, employees determine how much lower their pension payments would be if they quit (aka the quit pension) and compare it to the value of their future pension if they stayed (aka the stay pension).

Continuance Commitment And Ranked Results

The first test for the continuance commitment and pension cost framework was to compare the survey’s ranking results to the framework. I did this to determine if the results made sense under the framework. In my opinion, and as I wrote in my thesis, the framework appeared to explain most of the ranking results accurately, save one. For instance, six of the top eight ranked design elements are lucrative for employees and create stay pension value. The other two in the top eight, non-portability and no social security contributions, are penalties and impose high pension quit costs, which I also explained in Parts 25 and 26.

The only noted aberration from the framework was the non-contribution design element’s low ranking at number ten. It is both a lucrative design element and increasingly rare (Brainard & Brown, 2018). Under the framework’s line of reasoning, I expected much higher efficiency scores. Its low scores may reflect an acknowledgment of the wage offsets mentioned in Part 25 that accompany some non-contribution pensionable jobs (Ippolito, 1994; Montgomery et al., 1992). Or, alternatively, the survey participants who had access to it may not have realized that most pensionable employees must make mandatory contributions. In other words, they may not have known how rare or lucrative this feature really is.

Write-Ins

Another point worth making regarding the continuance commitment framework is that ten survey participants used the write-in option and stated that the money is what made them consider staying the most. Now, I could’ve combined those write-in answers into the basic annuity category because, as I pointed out in Part 25 of the Pension Series, every pension offers an annuity, and even a basic annuity holds a lot of stay pension value. However, because the write-ins didn’t expressly state that they stayed for the annuity money (meaning they could’ve stayed for a lump sum), I didn’t combine the two categories.

As a result, I created an ‘other’ category and placed those answers in it. That category came in at an eleventh place. That said, staying for the money is a high continuance commitment decision and another sign that the framework is a potentially useful explanatory tool for determining the impact of pension plan design on Golden Albatross decisions.

Another Continuance Commitment Data Point

By the way, I also tested to see if continuance commitment was present in the overall Golden Albatross decisions of the participants. I didn’t just assume that what I was seeing at the design element level of my pension survey was continuance commitment. So, at the end of the survey, I asked all 144 qualifying participants to judge the proportional size of their pension’s role in their reasons for staying during their stay-or-go decision. It started with a “0% to 20%” category and moved all the way to an “81% to 100%” category in 20% chunks. The chart below displays the results for the 144 respondents, 130 of whom also ranked their features.

If continuance commitment wasn’t a good explanatory tool, I would’ve expected to see either an even spread between the answers or a large number of low percentage scores. However, as the chart shows, nearly two-thirds of respondents signaled that their pension made up between 61% and 100% of the reason(s) why they considered staying. Moreover, over one-third of the respondents stated their pension made up 81% to 100%. These results were significant because the question measured the pull effect that DBPs create during a Golden Albatross decision, which the literature showed was likely due to continuance commitment (Luchak & Gelallty, 2001; Luchak et al., 2008).

Conclusion

There you have it. According to the ranking portion of my pension survey, the pension design element that made employees consider staying the most during their Golden Albatross decision was pension subsidized healthcare. Healthcare was followed by the immediate annuity and low-risk retirement income design elements at a distant second and third, respectively. Most of the top ten results made sense under the continuance commitment and pension quit cost framework discussed in Part 26 of the pension series. The only result that didn’t was the non-contribution pension design element. Of course, only participants who had access to the design elements voted for them. Therefore, comparative scores were created through efficiency calculations.

However, as much as the ranked results and the continuance commitment framework seem to complement each other, there’s still more analysis to come, and it tells a more convoluted story. In fact, statistical analysis shows there are far stronger forces at work than any singular pension design element’s impact through continuance commitment on a pensionable employee’s Golden Albatross decision. However, that’s the topic for the next post. Given that the ranking results convey participants’ stated preferences, they deserve consideration in their own right, which this article provides.

References (A – I)

Aubry, J.-P., Munnell, A. H., & Wandrei, K. (2020). 2020 update: Market decline worsens the outlook for public plans. Center for Retirement Research at Boston College. https://crr.bc.edu/briefs/2020-update-market-decline-worsens-the-outlook-for-public-plans/

Aubry, J.-P., & Wandrei, K. (2020). The status of local government pension plans in the midst of COVID-19. https://crr.bc.edu/briefs/the-status-of-local-government-pension-plans-in-the-midst-of-covid-19/

Aubry, J., Crawford, C. V, & Wandrei, K. (2018). Stability in Overall Pension Plan Funding. Center for Retirement Research at Boston College, 62 (October). https://crr.bc.edu/briefs/stability-in-overall-pension-plan-funding-masks-a-growing-divide/

Benartzi, S., Previtero, A., & Thaler, R. H. (2011). Annuitization puzzles. Journal of Economic Perspectives, 25(4), 143–164. https://doi.org/10.1257/jep.25.4.143

Brainard, K., & Brown, A. (2018). Spotlight on significant reforms to state retirement systems. In National Association of State Retirement Administrators (Vol. 32, Issue December). https://www.nasra.org/files/Spotlight/Significant Reforms.pdf

Enns, J. H., Nelson, G. R., & Warner, J. T. (1984). Retention and retirement: The case of the U.S. military. Policy Sciences, 17(2), 101–121. https://doi.org/10.1007/BF00146923

Hoffman, A. K., & Jackson, H. E. (2013). Retiree out-of-pocket healthcare spending: A study of consumer expectations and policy implications. American Journal of Law and Medicine, 39(1), 62–133. https://doi.org/10.1177/009885881303900102

Ippolito, R. A. (1991). Encouraging long-term tenure: Wage tilt or pensions? Industrial and Labor Relations Review, 44(3), 520–535. http://www.jstor.com/stable/2524159

Ippolito, R. A. (1994). Pensions and indenture premia. The Journal of Human Resources, 29(3), 795. https://doi.org/10.2307/146253

Ippolito, R. A. (2002). Stayers as ‘workers’ and ‘savers’: Toward reconciling the pension-quit literature. The Journal of Human Resources, 37(2), 275–308. http://www.jstor.com/stable/3069648

References (J – W)

Jennings, W. W., & Reichenstein, W. (2003). Valuing defined-benefit plans. Financial Services Review, 12(3). https://go.gale.com/ps/anonymous?id=GALE%7CA149157818&sid=googleScholar&v=2.1&it=r&linkaccess=abs&issn=10570810&p=AONE&sw=w

Jones, K. (2017). The most desirable employee benefits. Harvard Business Review. https://hbr.org/2017/02/the-most-desirable-employee-benefits

Liang,  Ian, Moore, B., & Soni, A. (2020, July). National inpatient hospital costs: the most expensive conditions by payer, 2017. Agency for Healthcare Research and Quality Statistical Brief #261. https://www.hcup-us.ahrq.gov/reports/statbriefs/sb261-Most-Expensive-Hospital-Conditions-2017.jsp

Luchak, A. A., & Gellatly, I. R. (2001). What kind of commitment does a final-earnings pension plan elicit? Relations Industrielles, 56(2), 394–417. https://doi.org/10.7202/000030ar

Luchak, A. A., & Gellatly, I. R. (2002). How pension accrual affects job satisfaction. Journal of Labor Research, 23(1), 145–162. https://doi.org/10.1007/s12122-002-1023-5

Luchak, A. A., Pohler, D. M., & Gellaty, I. R. (2008). When do committed employees retire? The effects of organizational commitment on retirement plans under a defined-benefit pension plan. Human Resource Management, 47(3), 581–599. https://doi.org/10.1002/hrm.20233

Montgomery, E., Shaw, K., & Benedict, M. E. (1992). Pensions and wages: An hedonic price theory approach. International Economic Review. https://doi.org/10.2307/2526986

Nyce, S. A. (2007). Behavioral effects of employer-sponsored retirement plans. Journal of Pension Economics and Finance, 6(3), 251–285. https://doi.org/10.1017/S1474747207003101

Sharp, L. M., & Biderman, A. D. (1966). The employment of retired military personnel. In Office of Manpower Policy, Evaluation and Research U.S. Department of Labor. https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&ved=2ahUKEwjd8rCwwP3vAhX34jgGHaXuDLAQFjAAegQIAhAD&url=https%3A%2F%2Fapps.dtic.mil%2Fdtic%2Ftr%2Ffulltext%2Fu2%2F646463.pdf&usg=AOvVaw349ZUPlAFjsT1jUQK-wTWz

Warner, J. T. (2008). Thinking about military retirement: An analysis for the 10th QRMC. https://www.cna.org/CNA_files/PDF/D0017798.A1.pdf