comparing union sponsored and private pension plans

7/28/2019 Comparing Union Sponsored and Private Pension Plans

http://slidepdf.com/reader/full/comparing-union-sponsored-and-private-pension-plans 1/58

Comparing

Union-Sponsored andPrivate Pension Plans:How Safe Are Workers’Retirements?

DIANA FURCHTGOTT-ROTH

AND ANDREW BROWN

Hudson Institute



Comparing

Union-Sponsored andPrivate Pension Plans:

How Safe Are

Workers’ Retirements?

Diana Furchtgott-RothAndrew Brown

Hudson InstituteSeptember 2009



© 2009 Hudson Institute

ISBN 978-1-55813-167-5

This report is an update o the Summer 2008 study, entitled Union vs. Private

Pension Plans: How Secure Are Union Members’ Retirements?

Hudson Institute is a nonpartisan policy research organization dedicated to

innovative research and analysis that promotes global security, prosperity, and

reedom. We challenge conventional thinking and help manage strategic tran-

sitions to the uture through interdisciplinary and collaborative studies in de-

ense, international relations, economics, culture, science, technology, and law.

Through publications, conerences, and policy recommendations, we seek toguide global leaders in government and business. For more inormation about

Hudson Institute, visit our website at www.hudson.org.

Diana Furchtgott-Roth is a senior ellow at the Hudson Institute. She directs

the Center or Employment Policy, whose purpose is to explore undamentally

sound economic policies that encourage economic growth and employment cre-

ation. She is the editor o Overcoming Barriers to Entrepreneurship in the Unit-

ed States (Lexington Books, 2008). From 2003 to 2005 Ms. Furchtgott-Roth

was chie economist at the U.S. Department o Labor. She received her B.A. in

economics rom Swarthmore College and her M.Phil. in economics rom Oxord

University.

Andrew Brown is an independent economist. He has provided research or

the Small Business Administration and the Hudson Institute. He was previously

a research assistant with the Hudson Institute Center or Employment Policy,

where he analyzed pension data or the 2008 paper Union vs. Private Pension

Plans: How Secure Are Union Members’ Retirements? He graduated with hon-

ors rom Swarthmore College with a B.A. in economics and mathematics.

The authors are grateul or the research assistance o Astha Shrestha. The

views expressed in the paper are those o the authors alone, and do not necessar-

ily refect those o any other party. The authors take responsibility or all errors.



Table o Contents

I. Introduction 1II. Executive Summary 1

III. What are Pensions? 4

IV. Trends in Pension Coverage 7

V. Sources o Inormation on Pension Plans 10

VI. General Health o Pension Plans in the United States 11

VII. Why Union Plans Tend to Perorm More Poorly 16

VIII. Small Plans 19

IX. Rank-and-File Versus Ofcer Pension Plans 21

X. Underunded Pensions and the Employee Free Choice Act 23

XI. Political Spending 25

XII. Eight Case Histories 28

Service Employees International Union 28

E.I. DuPont 29

International Brotherhood o Teamsters 30

Plumbers and Pipeftters 32

Southern Caliornia, Arizona, Colorado and Nevada Glaziers 33

United Auto Workers 34

Sheet Metal Workers 35

UNITE Here Fund Administrators, Inc. 37

XIII. Conclusions 39

Appendix I: How Do Pensions Work? 42

Appendix II: The Pension Protection Act 45

Appendix III: Data 46

Bibliography 47

Endnotes 50



List of Tables and Charts

Chart 1 – Number o Pension Plans over Time 7

Chart 2 – Number o Pension Plans Sponsored by Single Employers over Time 8

Chart 3 – Number o Multiemployer Pension Plans over Time 9

Chart 4 – Average Ratio o Assets to Liabilities or Large Defned Beneft Plans 11

Chart 5 – Percent o Large Defned Beneft Plans That Are Fully Funded 12

Chart 6 – Percent o Large Defned Beneft Plans In Endangered Condition 13

Chart 7 – Percent o Large Defned Beneft Plans in Critical Condition 13

Chart 8 – Percent o Underunded Plans Paying At Least Minimum Annual Charges 14

Chart 9 – Percent o Plans in Critical Condition Paying At Least Minimum Annual Charges 15

Chart 10 – Percent o Large Defned Beneft Pensions Making Additional Contributions Because o Funding Defciency 15

Chart 11 – Average Annual Payment To Correct Funding Defciency 16

Chart 12 – Percent o Large Single Employer Defned Beneft Plans That Are Fully Funded 17

Chart 13 – Percent o Small Defned Beneft Plans That Are Collectively Bargained 20

Chart 14 – Percent o Large Defned Beneft Plans That Are Collectively Bargained 20

Chart 15 – Average Number o Participants in Small Defned Beneft Pension Plans 21



S e p t e m b e r 2 0 0 9 1

pension plans have provisions to ensure that

they remain sufciently unded to pay the

generous retirement benefts that unions ad-

vertise as part o the union benefts package?

This paper oers explanatory background

inormation on pensions, including sources

o inormation, reporting and disclosure re-

quirements o Congress and the U.S. Depart-

ment o Labor, and data and analysis that il-

luminate the causes o the underlying problem

o underunding. The study goes on to ana-lyze the general health o pension plans in the

United States and, in particular, documents

eight case histories that illustrate issues in

pension unding. It also examines the role po-

litical interests may play in union membership

and pension unding, and it analyzes the link

between legislation now pending in Congress,

the Employee Free Choice Act, and under-

unding o pensions. The paper will present

evidence o the disparity between union andnon-union pension plans, and call on union

members to recognize their leaders’ responsi-

bility to close that gap.

With union membership down to

only 8 percent o private sector

workers—or 12 percent o em-

ployed wage and salary workers—unions are

doing their best to recruit new members. The

advertised benefts o joining a union sound

appealing: unions claim time and time again

that their members are more likely to have

health insurance, defned beneft pension

plans, and higher wages.

What unions do not tell prospective re-cruits is that there is a widespread pattern

o poor perormance among collectively-bar-

gained defned beneft pension plans. They

perorm poorly when compared to plans

sponsored unilaterally by single employers

or non-union employees. And pensions spon-

sored by unions that cover many employers

(multiemployer plans) are worse than pen-

sions oered by a single employer under a

collective bargaining agreement with a union.This disparity raises an important question:

is the promise o a union-run pension plan a

valuable one? To put it another way, do union

I. Introduction

II. Executive Summary

Unions recruit new members by claim-ing that their members are more

likely to have health insurance, de-

fned beneft pension plans, and higher wages.

However, an analysis o the fnancial status

o individual pension plans shows that col-

lectively bargained pension plans perorm

poorly when compared to plans sponsored

unilaterally by single employers or non-unionemployees.

Pensions are regular payments made to

retired workers rom money that they and

their employers put aside during their work-

ing years. Pensions come in two broad cate-

gories: defned beneft pensions, and defned

contribution pensions. A defned beneft pen-



H U D S O N I N S T I T U T E2

are more generous than can be aorded, o-

ten seeking pension increases in the ace o

consistent employer complaints about costs.

While union leaders have many incentives to

bargain or increased benefts, they have ewincentives to ensure that these benefts can be

easily paid or. However, this is not the case

with sta and ofcer pension plans.

Our analysis o 30 sta and ofcer pen-

sion plans relating to some o the largest 46

rank-and-fle pension plans shows that of-

cer pension plans may be better unded than

those or rank-and-fle. On average, the rank-

and-fle pension plans had 79 percent o undsneeded to cover their obligations. Nine o the

pensions were ully unded, while 24 were less

than 80 percent unded. This is in contrast

with the ofcer plans, which were 93 percent

unded on average. While nine o the pensions

were ully unded, eight were less than 80 per-

cent unded.

Labor unions have been ound to have

contributed more than $130 million to the

2008 Senate, House and presidential electioncycles. While there is nothing wrong with

unions seeking to inuence election cycles, it

is important to note where this money came

rom and who decided how to spend it. The

SEIU has been explicitly demanding that its

membership fnance its political agenda. Po-

litical spending, by law, is supposed to come

rom voluntary contributions, not the dues

that members are required to pay. I unionsdivert dues money to political activity, union

members may discover that their dues were

earmarked to support candidates they oppose.

Eight union pension unds are analyzed

as examples. The Plumbers’ board o trust-

ees has kept their contributions above normal

costs, but they have regularly been using cred-

sion promises a specifc monthly stipend or a

retiree’s lietime, calculated using the number

o years worked and some measure o work-

ers’ earnings over that time. A defned con-

tribution pension sets up personal investmentaccounts; typically, the employee can make

some choices about how the money in his ac-

count is invested.

Data or 2006, the latest ull year o data

available, show that union-negotiated pen-

sion plans are consistently worse than their

non-union counterparts. Among large plans –

plans with 100 or more participants – 35 per-

cent o non-union plans were ully unded, ascompared to 17 percent o ully unded union

plans. While a pension plan need not neces-

sarily be ully unded at any given time to be

stable, the Pension Protection Act o 2006

considers unds with less than 80 percent o

their needed assets to be in “endangered” sta-

tus. While 86 percent o non-union unds had

80 percent or more unds needed to pay ex-

pected costs, only 59 percent o union unds

met the unding threshold. Similarly, whileonly 1 percent o non-union plans had less

than 65 percent o required assets, also called

“critical” status in the 2006 Act, 13 percent

o union plans were critical.

The problems with collectively bargained

pension plans extend even to smaller plans.

O non-union plans, 61 percent were ully

unded, compared to 25 percent o union

plans. Both union and non-union small planswere about as likely as their larger counter-

parts to have unding ratios below 80 percent.

Fiteen percent o the small union plans were

in critical condition, more than twice the ra-

tio o small non-union plans.

Union leaders have contributed to this

problem by negotiating or pension plans that



S e p t e m b e r 2 0 0 9 3

orcing them to take account o their insecure

positions and to make eorts to rebuild since

2008.

However, the real problem is the opacity

o union pension fnancing and the lack o ac-

countability required o union leaders. Union

members have ew assurances that the trustees

o their pension unds are truly acting in their

best interest. The continued poor status o

union-run pension plans suggests that union

trustees are not adequately working towards

ensuring that rank-and-fle members have the

stable fnancial utures they deserve.

its to put o needed contributions towards

shortalls. Although the Sheet Metal Workers’

leadership successully lobbied or more ben-

efts beore 2008, they were not able to secure

them, leading to reductions in many “adjust-

able” benefts. The South Carolina, Arizona,

Colorado & Southern Nevada Glaziers, Ar-

chitectural Metal & Glass Workers pension

plan administrators made poor investment

choices, leading to investment losses.

Until the passage o the Pension Protection

Act o 2006, it was difcult to ensure that

contributions matched liabilities. The Act has

exposed more than 300 poorly unded plans,




Pensions come in two broad categories:

defned beneft pensions and defned contri-

bution pensions. A defned beneft pension

promises a specifc monthly stipend or a re-

tiree’s lietime. This sum is oten calculated

by using the number o years worked and

some measure o the worker’s earnings over

that time. The worker may or may not con-

tribute to the pension plans, but the employer

always contributes.

A defned contribution pension plan sets upan investment account or each worker. I it is

a “contributory” plan, the worker contributes

part o each paycheck—perhaps our or fve

percent—into the account. The employer may

make so-called “matching” contributions, al-

though the term is something o a misnomer

because the employer’s payment is oten less

than the worker’s, perhaps one-third or one-

hal. The money in the defned contribution

account, oten known as a 401(k) account,is invested and appreciates over time. How

much retirement income the defned contri-

bution account will generate depends on the

amounts invested, choice o asset and how

well it perorms, and years invested.

The sponsor o a defned beneft plan (usu-

ally an employer, but see below) is required to

make payments to retirees at the rate specifed

by the terms o the pension plan. The 1974Employee Retirement Income Security Act

(henceorth reerred to as ERISA) dictated that

sponsors must prepare to meet these uture

obligations by unding them beorehand—

contributing a certain amount o money per

participant every year into a pooled invest-

ment account.

Pensions are regular payments made to

retired workers rom money that they

and their employers put aside during

their working years. (In the United States, the

best-known pensions are the monthly Social

Security benefts that the ederal government

pays. However, they are not unded in the way

that employer pensions are.) As well as Social

Security, some workers have nongovernment

pensions earned during their years o employ-

ment in the private sector; others have pen-sions earned working or state, local or eder-

al government. In addition to these pensions,

many Americans have retirement assets in the

orm o tax-avored savings accounts, such as

Individual Retirement Accounts and 401(k)

plans, some o which have contributions rom

past and current employers. In recent decades,

there has been a steady decline in the number

(and proportion) o private employers that o-

er pensions that promise a specifc monthlyretirement beneft.

All types o pensions can be considered

deerred compensation— that is, a part o a

worker’s earnings not paid immediately. In an

age when workers were expected to work at

one company or most o their working lives,

this was a useul tool to encourage company

loyalty and get workers invested in staying

on or many years. For workers, the promiseo regular retirement income was attractive,

and so they would work or an employer or

decades, sometimes keeping a job that they

ound disagreeable or barely tolerable. Many

pensions are not legally owed to a worker un-

til he has worked at a company or several

years, when the pension becomes “vested.”

III. What are Pensions?



S e p t e m b e r 2 0 0 9 5

In contrast, participants in a defned con-

tribution plan have a legal claim on the money

they have contributed rom their paychecks to

personal accounts, as well as the investment

return on this money. Workers have some say

over how the money is invested—in bonds,

stocks, or a mixture. Moreover, defned con-

tribution plans are portable; workers take

them when they change jobs. This is an im-

portant consideration in an economy in which

lietime employment is becoming increasingly

uncommon. With portable accounts, workers

are ree to move to more attractive jobs. O

course, there are drawbacks to defned con-tribution plans. Without possibly costly in-

vestment advice, a worker may not have the

expertise to invest his or her account wisely.

And securities market downturns make the

worker’s retirement income vulnerable, es-

pecially i the worker retires when securities

prices are alling steeply, as in the winter o

2008-09.

Unions oten advocate defned beneft

plans in part because they preer employ-ers to bear the risk o supporting the rank-

and-fle’s retirement. However, as mentioned

above, defned beneft plans have the eect o

requiring workers to stay at unionized jobs in

the same industry or most o their working

lives, to maximize their retirement income.

Unions that negotiate defned beneft plans

create strong incentives or union workers not

to leave or non-union jobs and to continueto contribute to the defned beneft und. In

other words, defned beneft plans may con-

tribute to union security.

Another important distinction in pension

accounting is the dierences between “single-

employer,” “multiemployer,” and “multiple-

employer” pension plans. A single-employer

The sponsor is meant to invest this money

wisely in order to accumulate enough unds

to pay promised benefts when they come due.

This approach requires companies to predict

how much they will pay into the und in theuture. These calculations, especially the u-

ture value o present contributions, make

managing a defned beneft plan complicated.

A fnancial manager (usually an institution) is

hired to make investment decisions so that the

und will grow enough to meet the sponsor’s

uture pension obligations to its employees.

Defned contribution plans create more

predictable costs or employers, who, in col-lective bargaining contracts, normally agree

to make contributions o a certain size, or to

match employee contributions to a certain ex-

tent.

Upon retirement, holders o defned con-

tribution accounts can purchase “annui-

ties”—contractual promises by fnancial in-

stitutions, oten insurance companies, to pay

a certain amount o money every month or

the rest o the holders’ lives. Or retirees cansimply withdraw the money and consume it at

their own pace. I they buy an annuity, they

eectively transorm their defned contribu-

tion plans into defned beneft plans.

Some people preer defned beneft plans

because an employer is usually liable or the

specifc promised uture benefts. However,

there is usually a risk that a company will ail,

even a big, “blue chip” company like GeneralMotors, which went into bankruptcy in 2009.

Even i employers become bankrupt, the pen-

sion und cannot be used to pay o creditors.

Sometimes, retired workers can expect the

U.S. government’s Pension Beneft Guaranty

Corporation to pay their benefts, in part i

not in ull.




tives1. The union negotiates with each em-

ployer and the board to determine contribu-

tion levels, and the board determines the level

o benefts and manages the und.

Finally, a multiple-employer pension plan

is usually adopted by a parent company to

provide pensions or employees in its afli-

ated companies. Trade associations or other

groups o employers may also choose to orm

multiple-employer pension plans.

The rationale behind multiemployer and

multiple-employer pension plans is that they

allow frms to pool risk among several em-

ployers. Multiemployer pension plans allowworkers to keep their pensions i they change

jobs to another participating company. This

consolidates union pension contributions into

larger investment pools. Multiple-employ-

er pension plans allow parent companies to

transer covered workers between subsidiaries

with minimal paperwork.

plan is sponsored by one frm to support the

retirement o its workers. Single-employer

plans may be adopted unilaterally by private

employers or be created by negotiation with

labor unions. Collectively-bargained single-

employer pension plans are managed by the

employer. While sums to be contributed may

be negotiated, the employer is responsible or

the health o the und, and the employer (or

designated fduciary) makes all investment de-

cisions.

Multiemployer pension plans are created

and sponsored by a labor union in order to

provide retirement income or workers inseveral dierent places o employment. This

requires the union, the sponsor o the plan,

to negotiate with each employer to join and

contribute to the und. They are oten called

“Tat-Hartley plans,” and are managed by a

board o trustees, which consists o an equal

number o union and employer representa-



700,000

800,000

Number of Pension Plans over Time

100,000

200,000

300,000

400,000

500,000

,

Total Plans

Total DB Plans

Total DC Plans

1 9 7 5

1 9 7 7

1 9 7 9

1 9 8 1

1 9 8 3

1 9 8 5

1 9 8 7

1 9 8 9

1 9 9 1

1 9 9 3

1 9 9 5

1 9 9 7

1 9 9 9

2 0 0 1

2 0 0 3

2 0 0 5

Source: U.S. Department o Labor, “Private Pension Plan Bulletin Historical Tables and Graphs”

http://www.dol.gov/ebsa/pd/1975-2006historicaltables.pd

Chart 1

S e p t e m b e r 2 0 0 9 7

that employers were more generous with their

oerings during this period. But since 1983,

the number o defned beneft pension plans

has allen, while the number o defned con-

tribution plans has increased and stabilized at

about 650,000. (See Chart 1) These changes

were due to shits in the preerences o both

employees and employers.

First, it became more common or em-

ployees to hold several jobs over their lietime,

making defned contribution plans more at-tractive or their portability. Second, employ-

ers began to view the long-term costs and

liabilities o defned beneft pensions as un-

certain and undesirable, and they developed a

preerence or the predictable annual costs o

defned contribution plans.

In 1976, defned beneft plans accounted or

nearly one-third o the 359,980 pension

plans in existence in the United States. By

2006, about 7 percent o the 694,550 plans

registered were defned beneft pensions. De-

fned beneft plans paid benefts to 22 mil-

lion people in 2006, and covered another 20

million current workers. Over the past thirty

years, while the number o total pension plans

has nearly doubled, the number o defned

beneft plans has allen rom a 1983 high o 175,000 plans to 48,600.

These shits occurred or several reasons.

First, workers began to view pension plans as

an integral part o the compensation package.

The growth in both defned beneft and de-

fned contribution plans until 1983 indicates

IV. Trends in Pension Coverage



Source: U.S. Department o Labor, “Private Pension Plan Bulletin Historical Tables and Graphs”

http://www.dol.gov/ebsa/pd/1975-2006historicaltables.pd

600,000

700,000

800,000

Number of Pension Plans Sponsoredby Single Employers over Time

0

100,000200,000

300,000

400,000

500,000

7 5

7 7

7 9

8 1

8 3

8 5

8 7

8 9

9 1

9 3

9 5

9 7

9 9

0 1

0 3

0 5

Total Plans

DB Plans

DC Plans

1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2

Chart 2


In 2006, collectively-bargained defned

beneft plans held approximately $890 billion

in assets. Multiemployer defned beneft plans,

most o them collectively-bargained, had ap-

proximately $340 billion. In light o the vast,acknowledged defciencies in union pension

unding, these assets should be much high-

er. At last count, 157 collectively-bargained

defned beneft plans had reported being in

“critical” status, that is, holding less than 65

percent o the assets needed to pay uture ob-

ligations. Another 146 were in “endangered”

status, generally reecting unds with less

than 80 percent o what was needed2

.Under the Pension Protection Act 2006,

defcient plans are required not only to report

signifcant unding inadequacies but to make

them up with reduced benefts or increased

contributions .

Single-employer plans, especially, have

driven these shits because most plans cover-

ing many frms are multiemployer plans, and

tend to be union-run. Unions avor defned

beneft plans, even i they do sometimes ne-gotiate the creation o a defned contribution

plan. Chart 2 shows that single employers

have strongly embraced the defned contribu-

tion plan as the preerred method o paying

deerred compensation.

Chart 3 demonstrates how these trends

have aected multiemployer pensions. While

it is difcult or a single employer, much less

several, to separate themselves rom a collec-tively-bargained agreement, multiemployer

pension plans have also begun to shit to-

wards defned contribution plans. In 2006,

over hal o the 3,037 multiemployer pension

plans were defned contribution plans.



Source: U.S. Department of Labor, “Private Pension Plan Bulletin Historical Tables and Graphs”

http://www.dol.gov/ebsa/pdf/1975-2006historicaltables

3,500

Number of MultemployerPension Plans over Time

500

1,000

1,500

2,000

2,500

3,000

Total Plans

DB Plans

DC Plans

0

1 9 7

5

1 9 7

7

1 9 7

9

1 9 8

1

1 9 8

3

1 9 8

5

1 9 8

7

1 9 8

9

1 9 9

1

1 9 9

3

1 9 9

5

1 9 9

7

1 9 9

9

2 0 0

1

2 0 0

3

2 0 0

5

Chart 3

S e p t e m b e r 2 0 0 9 9



H U D S O N I N S T I T U T E1 0

V. Sources of Informationon Pension Plans

Most pension unds must fle Form

5500 annually with the Internal

Revenue Service and the U.S. De-

partment o Labor. Form 5500 includes in-

ormation about the assets and liabilities o

the pension und; the number o participants;

some details about the pension plan; and the

investment earnings o the pension und.

Form 5500 requires estimates o the val-ue o unds’ assets, liabilities, and the pres-

ent value o all beneft payments. Because o

the complexity o the calculation, companies

are asked to calculate only their “accrued”

liabilities. To do that they estimate the pres-

ent value o all benefts they would have to

pay i they closed at the end o the year and

paid all promised benefts based on service to

that time. That is, i benefts are 1 percent o

wages per year o service, a person who hadworked at the company or 10 years would be

owed an annuity o 10 percent o his wages

each year.

I the ratio o assets divided by liabilities

is greater than or equal to one, the company

would be able to pay all o its promised ben-

efts assuming all actuarial assumptions are

correct. This is an important qualifcation

and will be discussed later. I this ratio is lessthan one, it indicates that the company has

promised to pay more in benefts than it ex-

pects to have when benefts are paid. A ratio

o less than one is the major indication to plan

participants that their pension und—and

possibly their own benefts—are in danger. A

ratio less than one, an “underunded” plan,

reects mismanagement o unds, either inad-

vertent or corrupt.

The Employee Retirement Income Secu-

rity Act o 1974 (ERISA), a law written to

protect defned beneft pension plans, permits

sponsors to delay required contributions, as

long as they adhere to payback restrictions.

However, another provision o ERISA, in-tended to encourage pension sponsors to keep

their plans well-unded, has contributed to

the deterioration o unds.

In any year in which a sponsor pays more

than its required contributions, ERISA allows

the und to gain a “credit” that can be used to

reduce uture payments. That happened dur-

ing the “dot-com bubble” o the late 1990s,

when many pension unds ran up large cred-

its due to unexpected appreciation o theirstock-market assets. Sponsors substituted the

credits or annual contributions. But when the

bubble burst and stocks ell, the unds started

perorming poorly. Then und administrators

used the credits to pay o the debts they ac-

cumulated during the market’s decline, and at

times did not make any contributions to their

pensions, even i their unded ratios had allen

signifcantly.The data used in this study were drawn

rom the Employee Benefts Security Admin-

istration (EBSA) Research File, compiled by

the Center or Retirement Research at Boston

College. We would like to thank the Center

or making these data available or this paper.



Source: U.S. Department o Labor Employee Beneft Security Administration Form 5500 Data

(2006), Center or Retirement Research at Boston College. 5500-CRR data: Panel o Current and

Usable Form 5500 Data. Chestnut Hill, MA, and Hudson Institute calculations.

Average Rao of Assets to LiabiliesFor Large Defined Benefit Plans

97%

0

20

40

60

80

100

Non-Union Plans

P e r c e n t o f L i a b i l i e s C o v e r e d

86%

Collecvely-Bargained Plans

Chart 4

S e p t e m b e r 2 0 0 9 1 1

counterparts, 17 percent o which were ully

unded (Chart 5). (“Fully unded” means that

assets meet or exceed 100 percent o the value

deemed necessary to pay all projected uture

obligations.)

The 2006 Pension Protection Act, like

ERISA (1974), recognized that a pension plan

need not be ully unded at any moment in

time to be stable. Dips in the stock market

can reduce the unding ratio, requiring sever-al years or recovery, even among well-unded

pensions. Thus, while und managers must

make up shortalls over the course o several

years, there are ew penalties i unding alls

to or below 80 percent o projected liabilities.

At the end o 2006, large defned ben-

eft pension plans3 were, on average,

92 percent unded4. The data came

rom 4,602 large non-union pension plans

and 3,481 collectively-bargained pension

plans. (The threshold or “large” was 100 or

more participants.)

Collectively-bargained large plans, on av-

erage, were 86 percent unded. In contrast,

non-union pensions were 97 percent unded(Chart 4). This suggests systemic underund-

ing among collectively-bargained pension

plans. Non-union plans, 35 percent o which

were ully unded, were more likely to be

ully unded than their collectively-bargained

VI. General Health of Pension Plansin the United States



Source: U.S. Department o Labor Employee Beneft Security Administration Form 5500 Data (2006),

Center or Retirement Research at Boston College. 5500-CRR data: Panel o Current and Usable

Form 5500 Data. Chestnut Hill, MA, and Hudson Institute calculations.

Percent of Large Defined Benefit PlansThat Are Fully Funded

35%

0

10

20

30

40

Non-Union Plans

P e r c e n t o f D e fi n e d

B e n e fi t P l a n s

17%


Chart 5


out incentives to discontinue this practice, it is

possible that many plans relied on large und-ing credits to keep rom having to make sup-

plemental contributions even when the stock

market ell, pulling down unding ratios.

One way o lending strength to this theory

is analyzing whether plans that were not ully

unded received contributions equal to total

annual charges, generally equal to the sum o

the normal cost (see Appendix) and interest

charges rom unding defciencies. In 2006,only 19 percent o the 2,882 union pension

plans that were not ully unded, 548 plans,

made contributions sufcient to cover their

minimum annual costs. In contrast, 1,079, or

37 percent o the 2,944 non-union pension

plans that were not ully unded, met annual

costs ( Chart 8 ).

However, in 2006, collectively-bargained

pension plans still perormed poorly com-pared to non-union pensions. O union plans,

41 percent had less than 80 percent o their

needed assets, compared to 14 percent o non-

union plans, shown in Chart 6.

Unsurprisingly, the pattern repeats itsel

among the most troubled plans – those with 65

percent or less o their required assets, called

“critical status” in the 2006 Pension Protec-

tion Act

5

. While only 1 percent o non-unionpensions were in critical status, 13 percent o

union pension plans were critical (Chart 7).

One possible reason or the superior per-

ormance o non-union plans is the use o

unding credits to reduce mandatory pay-

ments to pension plans, a practice discussed

above. While the Pension Protection Act held




(2006), Center or Retirement Research at Boston College. 5500-CRR data: Panel o Current

and Usable Form 5500 Data. Chestnut Hill, MA, and Hudson Institute calculations




Percent of Large Defined Benefit PlansIn Endangered Condion

14%

0

10

20

30

40

Non-Union Plans



41%


Percent of Large Defined Benefit PlansIn Crical Condion

1%0

5

10

15

Non-Union Plans



13%


Chart 6

Chart 7

S e p t e m b e r 2 0 0 9 1 3



Percent of Underfunded PlansPaying At Least Minimum Annual Charges

37%

0

10

20

30

40

Non-Union Plans



19%

Collectvely-Bargained Plans

Chart 8





The situation was worse or plans in

critical condition. Among 438 union pen-sion plans in critical condition, only 24, or 5

percent, contributed enough to meet annual

costs. O the 54 non-union plans in critical

condition, 21, or 39 percent, were in the same

situation (Chart 9).

Under the 1974 law, some pension

plans in poor condition would have to

make additional contributions. In 2006,

19 percent o non-union plans had to makeadditional contributions, and 20 percent

o union plans had to make such contri-

butions (Chart 10). However, on average,

union plans were orced to make contri-

butions close to twice the size o those o

non-union plans, $4.9 million compared to

$2.5 million (Chart 11).

Compared to 2005, pensions in 2006

ared worse as a whole. Union pensions were88 percent unded in 2005, and non-union

pensions 98 percent unded on average, mak-

ing their 2006 showings small but noticeable

losses. Both union and non-union pension

plans saw a 2 percentage point drop in the

number o pensions that are ully unded.

Non-union pensions were hal as likely to be

in critical condition in 2006 as in 2005, while

the 13 percent o critical union pensions were2 points higher than 2005 union pensions.

One would expect, in a year o stock mar-

ket gains, or pensions to improve in health.

It is thereore distressing to see their health

deteriorate, and especially or the increased

likelihood or union pensions to be in critical

status.









Percent of Plans In Crical CondionPaying At Least Minimum Annual Charges

39%

0

10

20

30

40

Non-Union Plans



5%


Percent of Large Defined Benefit PensionsMaking Addional ContribuonsBecause of Funding Deficiency

19%

0

5

10

15

20

Non-Union Plans

P e r c e n t o f F u n d s

20%


Chart 9

Chart 10

S e p t e m b e r 2 0 0 9 1 5



Source: U.S. Department o Labor Employee Beneft Security Administration Form 5500 Data (2006),

Center or Retirement Research at Boston College. 5500-CRR data: Panel o Current and Usable Form

5500 Data. Chestnut Hill, MA, and Hudson Institute calculations.

Average Annual PaymentTo Correct Funding Deficiency

2.5

0

2

1

3

4

5

Non-Union Plans

M i l l i o n s o f D o l l a r s

4.9


Chart 11


negotiate with several dierent employers

(which is the case when a union sponsors a

multiemployer pension plan), this problem

may be exacerbated. The data suggest thisexplanation has merit. O multiemployer pen-

sions, 6 percent were ully unded in 2006. In

contrast, 33 percent o multiple employer and

31 percent o single employer pensions were

ully unded.

Unortunately or this hypothesis, the dis-

parity between union and non-union plans

As noted above, union pension plans

are more commonly underunded

than are non-union plans sponsored

unilaterally by employers. One possible rea-son or the disparity is that collectively-bar-

gained pension plans are not usually renewed

annually. As a result, annual contributions

by employers (and possibly employees) may

not respond quickly to market downturns

or other unexpected drops in pension und-

ing ratios. Furthermore, when a union must

VII. Why Union Plans Tend to Perform More Poorly



Source: U.S. Department o Labor Employee Beneft Security Administration Form 5500 Data (2006), Cen-

ter or Retirement Research at Boston College. 5500-CRR data: Panel o Current and Usable Form 5500

Data. Chestnut Hill, MA, and Hudson Institute calculations.

Percent of Large Single Employer DefinedBenefit Plans That Are Fully Funded

35%

0

10

20

30

40

Non-Union Plans



23%


Chart 12

S e p t e m b e r 2 0 0 9 1 7

bers when they renew collective contracts. It

makes their re-election more likely. But lean-

ing on employers to ensure that the pension

plan is kept well-unded takes much work

or little visible eect—and may well work

against winning richer benefts by underscor-

ing their cost to the employer. It sounds much

more proactive or union leaders to deliver ex-

panded pension benefts or to protect benefts

rom employer-initiated cuts than to protect

already-earned benefts.

exists even within dierent employer groups.

Among single employer plans, 35 percent o

non-union plans were ully unded, compared

to 23 percent o single employer union plans.

(Chart 12)

There are other possible explanations or

the poorer showing made by union pensions.

One involves the incentives o union ofcers,

what might be called internal trade union

politics. Union leaders like to achieve ex-

panded uture pension benefts or their mem-




ree index o pension health. While optimism

is commendable, union leaders’ tendency to

project rosy utures or pension plans leaves

rank-and-fle members unaware when or how

their pensions may be at risk. And without

that understanding, union members have no

metric by which to judge the management by

their leaders o their pension plan.

This is a problem because a secure retire-

ment is one o the primary enticements to join

a union. Many o the major national unions

advertise that union members are more likely

to have better retirement plans, explicitly de-

fned beneft retirement plans, than workerswithout a union. This promise would be ar

less persuasive were potential union members

to understand that a pension is not guaran-

teed just because it has been created.

In act, looking at union communications,

it is clear that union leadership prioritizes

raising benefts over securing them. Nowhere

is this clearer than in criticisms o the Pension

Protection Act. The Teamsters, or example,

criticize ull unding requirements as this pro-vides a legal reason or employers to reuse to

increase pension benefts6. Unions requently

lambaste their employer opponents or op-

posing increased beneft plans, usually listing

their proposals as “reasonable” or “aord-

able.” But it is rarely in a union’s interest to

say that it is pushing or reorms o uncertain

cost that employers may not be able to aord.

The act that unions oten push or beneftincreases in the ace o employer protests o

unreasonable cost lends weight to the argu-

ment that they place more importance on the

promised level o benefts than on the actual

security o those benefts.

There are advantages to defned beneft

pension plans, but these benefts are contin-

Union leaders also may not want to ad-

vertise to members that saeguarding pension

plan unding requires eort. It is part o trade

union doctrine that defned beneft plans are

sae. For the leadership to acknowledge that

it must struggle to sustain adequate unding

suggests that the annual contributions are in-

adequate to assure rank-and-fle members a

stable retirement. The implication that larger

contributions are needed may apply to em-

ployees, i the plan is contributory, as well as

to the employer. Or it might imply that the

plan should become contributory. It might

even be inerred by the membership that theleadership has bargained or gold-plated ben-

efts that are unrealistic. In sum, conronting

under-unding is a headache or union lead-

ers and the path o least resistance is to avoid

such engagement.

Finally, employers may be more disposed

to grant pension beneft increases than wage

increases because the ormer are largely uture

costs. Union leaders understand that and may

tilt their demands in that direction, especiallyi the employer is in straitened circumstances

and cannot aord to raise wages.

It should be noted frst that all o these

reasons listed are hypotheses. There appears

to be some evidence to support them, and

they are plausible given the current environ-

ment, but we cannot actually determine what

drives a union leader’s decisions.

But while not all o these proposed causes

or disparity between non-union and collec-

tively-bargained pension plans are due to out-

right malice or incompetence, they still merit

concern. Annual reports are as a rule optimis-

tic about the uture o a pension plan, oten us-

ing the plan’s accumulated assets as a context-



S e p t e m b e r 2 0 0 9 1 9

o small pensions may be more likely to know

and have contact with their union representa-

tive, giving them greater opportunity to seek

and maintain accountability. One can, there-

ore, expect small plans in general to do bet-

ter than larger pensions.

O non-union small plans, 61 percent were

ully unded, compared to 25 percent o union

plans. Both sets outperormed large plans in

this respect, but, taking account o underper-

orming unds, neither did well. Both union

and non-union small plans were about as like-

ly as their larger counterparts to have und-

ing ratios below 80 percent. Small and large

union plans were nearly three times as likely

to be poorly unded as were non-union pen-

sions. Worse yet, 15 percent o small union

plans had unding ratios below 65 percent—

more than twice the ratio or small non-union

plans.

gent on the plans being ully unded. Union

rank-and-fle members need to understand

the complexity o defned beneft pension

plans, and hold their leaders responsible or

maintaining stable unding ratios. Certainly it

is in employers’ best interests to rerain rom

increasing benefts during negotiations, but

they also have a stake in ensuring the benefts

they pay are sustainable. Short-term business

gains, or example, are no justifcation or in-

creasing benefts, in part because it is much

harder or a business to rescind increased ben-

efts during down times than to not increase

them in the frst place.

But more important than acknowledging

the possibility that employers may at times be

accurate about the easibility o union beneft

plans, rank-and-fle workers must recognize

the difculty in keeping a defned beneft pen-

sion stable, and properly place responsibility

or that task. Employers are responsible or

making the contributions stipulated in their

contracts, and, since they also serve as trust-

ees, they need to make sure that the plan is

stable. It is the union leaders’ responsibility to

negotiate stipulations that enable the pension

plans to remain stable.

VIII. Small Plans

Like businesses, pension plans are oten

analyzed by size. Small plans, those

covering 100 or ewer people, make

up the majority o defned beneft pension

in the United States, or 25,720 out o a totalo 33,803. Labor Department data rom the

2006 Forms 5500 showed that o these small

plans7, 827, or 3 percent, were collectively bar-

gained, as opposed to the 43 percent o large

pensions that are collectively bargained (See

Chart 13 and Chart 14). These 827 union

plans were, on average, larger than non-union

pension plans. Small union plans covered on

average 69 workers, while non-union pen-sions covered an average o 13 people apiece

(Chart 15).

Small plans enjoy a natural advantage with

respect to unding. The ewer people the plan

must cover, the more likely it can accurately

predict uture obligations and adequately pre-

pare to meet them. Furthermore, benefciaries



Source: U.S. Department o Labor Employee Beneft Security Administration Form 5500

Data (2006), and Hudson Institute calculations

Source: U.S. Department o Labor Employee Beneft Security Administration Form 5500

Data (2006), and Hudson Institute calculations

Percent of Small Defined Benefit PlansThat Are Collecvely Bargained

3%

0

1

2

3

P e r c e n t o f P l a n s

Percent of Large Defined Benefit PlansThat Are Collecvely Bargained

43%

0

10

20

30

40

P e r c e n t o f P l a n s

Chart 13

Chart 14




Average Number of Parcipants inSmall Defined Benefit Pension Plans

12.6

0

20

40

60

80

Non-Union Plans

N u m b e r o f P a r c i p a n t s

68.6


Chart 15


(2006), and Hudson Institute calculations

S e p t e m b e r 2 0 0 9 2 1

How do these pension plans or union

leadership and sta compare to the pension

plans or the rank-and-fle?

It is not possible to perorm an analysis o

all sta pension plans. While Form 5500 pro-

vides inormation to indicate whether a plan

is collectively bargained or not, and indicates

One would hope that smaller union plans

would be better o than those run by large,

national unions, but it appears that the en-demic problems with collectively-bargained

pension plans extend even to smaller plans.

A local union negotiating a pension or 100

workers ought to be easy to hold accountable

or plan perormance, but the data suggest it

is not so. Even small plans are ar too otenin unstable fnancial condition, to the possible

uture detriment o the covered workers.

IX. Rank-and-FileVersus Ofcer Pension Plans

Many unions have separate pension

plans or rank-and-fle members

and or the sta and ofcers o

the union local and national organization,

because many union ofcers and sta are em-

ployees o the union, not o an employer with

whom the union bargains.




unded. Eleven o those 24 were less than 65

percent unded, and our were listed by the

Department o Labor as being in critical con-

dition.

The 30 ofcer unds were on average 93

percent unded. Nine were ully unded, and

eight were less than 80 percent unded. O the

eight, two were less than 65 percent unded.

These data suggest that sta pensions may be

better unded than rank-and-fle pensions.

There are two hypotheses already dis-

cussed that oer an explanation or the dis-

parate unding rates o rank-and-fle and of-

cer pension plans. Most ofcer pension plansare perks o the job, not collectively bargained

but dictated by the union’s bylaws. Further-

more, they are single employer pension plans.

Both o these distinctions have biases towards

better unding.

It is enlightening to observe, when looking

at the rank-and-fle plans, that o the 46, 17

are single-employer or multiple-employer pen-

sion plans. These plans are, on average, 96

percent unded, compared to 70 percent und-ing among the multiemployer rank-and-fle

union plans. Nine o these are ully unded,

and only our are less than 80 percent und-

ed. None are in critical condition. While these

statistics can not necessarily be generalized to

the rest o the pension universe, they do indi-

cate that ofcer pension plans are about as

well as single-employer rank-and-fle union

pension plans. This suggests that ofcer pen-sion plans are better than rank-and-fle plans

in part due to the ease with which ofcers can

manage their own pensions.

Union ofcers make many o the business

decisions or the union. They, or someone

they know, are aware o the fnancial status

o the union, and thereore what pension

whether a plan is large or small, ofcer pen-

sion plans cannot be easily distinguished rom

rank-and-fle plans. Few unions provide easily

accessible inormation on the pension plans o

their ofcers, and there is no central resourceor locating and examining them.

However, it is possible to analyze a sample

o sta pension plans, and this is the approach

taken in this study. We examined 30 sta

pension plans relating to some o the largest

46 rank-and-fle pension plans. Rank-and-fle

plans cover employees in the International

Brotherhood o Teamsters, SEIU, UNITE

HERE, UAW, the Communications Workerso America, IAMAW, the Sheet Metal Work-

ers, the United Steelworkers, the UFCW, the

Plumbers and Pipeftters, the Actors’ Equity

Association, the International Union o Paint-

ers and Allied Trades, the AFL-CIO, the Car-

penters, LIUNA, the International Union o

Bricklayers, the Bakery, Conectionery, To-

bacco Workers and Grain Millers, and the In-

ternational Brotherhood o Boilermakers.

Sta pension plans include afliates o na-tional unions such as AFSCME, the Ameri-

can Postal Workers8 and the Graphic Coner-

ence o the Teamsters. Rank-and-fle pensions

or these unions are consolidated in their par-

ent union’s pension. In addition, they include

the International Association o Iron Work-

ers, the International Association o Bridge

Workers, and the International Brotherhood

o Electrical Workers

9

.In 2006, rank-and-fle plans each covered

at least 25,000 active workers, and together

covered over 2.7 million active employees.

On average, the rank-and-fle plans had 79

percent o the unds needed to satisy their

obligations. Nine o the plans were ully

unded, and 24 were less than 80 percent



S e p t e m b e r 2 0 0 9 2 3

least as the bill was drated, would have the

power to orce newly-unionized frms into un-

derunded pensions. A union might proposesuch an assignment to pump up a plan’s und-

ing ratio. Changes in the bill’s language that

Congress should consider would resolve that

problem.

This legislation originally attracted much

attention because it would have allowed

workers to designate a union as their bargain-

benefts are aordable. This is similar to the

reason that single employer plans are better

than multiemployer pensions. That is, when

a single entity is responsible or determining

pension benefts, its sole responsibility or

benefts and its exibility allow it to better

keep the pension well-unded.

However, unlike a business with a collec-

tively-bargained pension plan, union ofcers

are not constrained by shareholders to keep

large pension defcits rom arising. They are

motivated by sel-interest: they are managing

their own pensions. That gives the ofcers

a greater incentive to ensure that their ownpensions are managed well than they have or

keeping their members’ pensions well unded.

It is not clear that they expend more eort

protecting their own pensions, but the out-

comes suggest that union leaders are more e-

ective in protecting their own utures.

Pay-or-perormance is one o organized

labor’s avorite tools or criticizing manage-

ment. One o their complaints is that among

large corporations, ew incentives exist or

upper management to increase profts. Or-

ganized labor argues that golden parachutes,

vested stock options, and other complex orms

o compensation ensure that most CEOs and

other senior corporate ofcers make dozenso times the average worker’s salary, even in

periods o declining profts or mass layos.

Union leaders say that business leadership

should not be rewarded when ordinary work-

ers are suering.

Union leaders, however, are vulnerable to

the same criticism. Ofcer pensions are rarely

connected to perormance o rank-and-fle

pension plans; ew union leaders’ uture in-comes are threatened when an employer must

cut pensions, or plans all into critical condi-

tion. Note that the 2006 slip in union pen-

sion unding occurred during a year o stock-

market gains. I unions wish to encourage

managers to work in the best interests o their

companies by connecting perormance con-

nected to profts, perhaps the unions ought to

set that example frst.

T

he sorry state o union pensions is one

reason that unions are embracing the

Employee Free Choice Act, EFCA, abill pending in Congress.

I enacted, the bill would require a union

newly certifed as a bargaining agent and the

employer that cannot agree on a contract

within 120 days, to submit to mandatory in-

terest arbitration. Resulting contracts would,

by law, hold or two years. The arbitrators, at

X. Underfunded Pensionsand the Employee Free Choice Act




that they must accept, possibly including re-

quiring frms to join underunded multiem-

ployer pension unds. The bill, as pending in

mid-2009, did not speciy eligibility criteria

or appointment to arbitration boards.

Unlike voluntary arbitration, the parties

would not have an opportunity to choose

members o the panel. Nor would they have

recourse to the courts i they were dissatis-

fed with the arbitrators’ decision. With just

a ew lines o legislative language, Congress

would revoke or newly-organized frms and

workers a critical principle o ree collective

bargaining—that employers and unions maywalk away rom a proposed contract they fnd

unsatisactory. Workers could be required to

accept a lower compensation package than

they could get elsewhere, and frms could

be orced into unproductive agreements that

could eventually lead to bankruptcy.

Mandatory interest arbitration would al-

low organized labor to replenish the coers

o its underunded pension plans. That’s why

it has been endorsed by union pension undmanagers. On May 11, 2009, a group o these

investors, including representatives o the

AFL-CIO and the Service Employees Interna-

tional Union pension plans, endorsed EFCA

in a letter to two senior Democrats, Senator

Tom Harkin o Iowa, cosponsor o the Em-

ployee Free Choice Act, and Representative

George Miller o Caliornia, chairman o the

House Education and Labor Committee. “Asfduciaries with broadly-diversifed porto-

lios,” the signatories wrote, “we must be cog-

nizant o these trends and their impact on our

investments.” 10

I an arbitration panel were to require a

frm to join an underunded plan, the frm

could become liable or the pensions o work-

ing agent by checking a card circulated and

retained by the union—hence the name “card

check”—rather than through a secret-ballot

election, as required since the Tat-Hartley

Act. Card check ran into intense opposition

in 2009 because it opened the possibility o

union intimidation. Union ofcers can visit

workers outside the plant or at home, and

they will know who signed and who reused.

(Interestingly, the unions assert that they have

lost secret-ballot elections because o intimi-

dation by management.) Because several Sen-

ate Democrats oppose card check, including

Blanche Lincoln and Mark Pryor o Arkan-sas, Diane Feinstein o Caliornia, and Ar-

len Specter o Pennsylvania, it has reportedly

been dropped rom the bill.

However, the mandatory interest arbitra-

tion part o the bill would remain in any alter-

nate version o EFCA. Unions want manda-

tory interest arbitration because they believe

that the threat o arbitration, ollowed by the

arbitration itsel, will orce employers to o-

er better compensation packages in the initialbargaining. Employers oppose interest arbitra-

tion on the grounds that it could require them

to pay excessive compensation to workers and

impose upon them onerous union work rules,

eventually orcing them out o business. The

contrast between the fnancial position o the

unionized Big Three Detroit auto companies

and the non-unionized oreign-owned trans-

plants shows how unionization aects a frm’scompetitive position.

Binding interest arbitration could have

even more pernicious consequences than end-

ing the secret ballot. It would allow a govern-

ment arbitration board, appointed by the Fed-

eral Mediation and Conciliation Service, to set

compensation packages or frms and workers



S e p t e m b e r 2 0 0 9 2 5

ingly Democratic, and their campaign contri-

butions reected this preerence.

According to the Center or Responsive

Politics, an organization that compiles Feder-

al Election Commission (FEC) data on cam-

paign contributions, labor unions contributed

more than $130 million to the Senate, House,and presidential races. There are several di-

erent ways or a union to inuence an elec-

tion. One is to fle their union under Section

527 o the U.S. Code, labeling their organiza-

tion a 527 group. Such groups are allowed to

make contributions to inuence national and

local elections, so long as the unding does not

ers, some already retired, o other employers.

This would generate an inow o new cash to

the plan but might harm the fnancial position

o the newly organized frm.

Under EFCA, i a trucking company wereunionized by the Teamsters and could not

reach an agreement with the union, the case

would go to mandatory arbitration. Arbitra-

tors could require the company to partici-

pate in a Teamsters pension und, such as the

Central States Pension Fund, which was 46.6

percent unded in 2006 and is used by many

trucking companies.

Under a multiemployer pension und suchas the Central States und, i some contribu-

tors go out o business then others have to

pay the obligations. This concept is known as

“last man standing.” Only i all the compa-

nies go out o business will the Pension Bene-

ft Guaranty Corporation, a government pen-

sion insurance und, pay retirees a maximum

beneft, now $12,870 or 30 years o service.

With ewer workers joining unions, the

collectively-bargained multiemployer pen-

sion unds are characterized by an increasingnumber o retirees supported by ewer young-

er workers. Many poorly unded pensions are

similar to Ponzi schemes, with new contri-

butions paid out in benefts rather than be-

ing saved or contributors’ retirement. Union

pension unds can survive only through new

contributions. That is why unions will do

anything to recruit new members —includ-

ing orcing workers into underunded pen-sion plans through mandatory arbitration.

But just as workers deserve secret ballots in

union elections, they also deserve the right to

consider judiciously their labor contracts, and

walk away rom those that they deem unat-

tractive or unair.

Rather than ocusing on unding pen-

sions or rank-and-fle workers, some

unions have concentrated their eorts

on politics. For example, in the 2008 elections,

many organizations poured billions o dollars

into the House, Senate, and presidential races,

and many unions were quick to back BarackObama. With his promises o supporting leg-

islation such as the Employee Free Choice Act,

universal health care, and other union agenda

items, he was a clear choice or union leaders.

But he was not necessarily a clear choice or

union members. Certainly, the unions would

have liked their members to vote overwhelm-

XI. Political Spending




ary Clinton’s bid. The American Federation

o Teachers spent $4.5 million trying to inu-

ence their members’ votes.

There is nothing inherently wrong with

unions seeking to inuence national elections.The First Amendment guarantees their right

to express their opinion, and unions arguably

have a duty to attempt to sway national poli-

tics in ways that beneft their members.

But looking at SEIU’s contributions o

more than $70 million to inuence 2008 elec-

tions14, or example, one wonders where this

money came rom and who decided how to

spend it. The SEIU Committee on PoliticalEducation (COPE) is unded by SEIU. While

many SEIU members were willing contribu-

tors, the 2008 SEIU constitution tells a dier-

ent story, namely that money to pay or po-

litical “education” does not come only rom

voluntary contributions. In 2008, the ollow-

ing amendment to the SEIU constitution was

adopted:

“Every U.S. Local Union shall contribute

an annual amount equivalent to at least $6.00per member per year, or as determined an-

nually by the International Executive Board,

to support the overall SEIU political educa-

tion and action program…I a Local Union

ails to meet its annual SEIU COPE undrais-

ing obligation, it shall contribute an amount

in local union unds equal to the defciency

plus 50%, or such other amount determined

by the International Executive Board, to sup-port the overall SEIU political education and

action program.”15

When the policy was highlighted in the

press, the union admitted that discipline like

this had always been in place, and that the

amendment had simply ormalized it.16 The

union may suggest that locals encourage vol-

directly advocate one particular candidate.

Aside rom some unions, 527 groups also in-

clude “issue” groups, who seek to inuence

elections based on diering positions on vital

issues, such as abortion rights, gun control,and, o course, labor issues.

Exempt rom many contribution and dis-

closure laws, 527 organizations across the po-

litical spectrum have been criticized or their

lack o accountability. Union-sponsored 527

groups contributed close to $57 million in the

2008 election cycle. The Service Employees

International Union contributed the most,

close to $28 million11

. Seven other union 527groups were among the top 50 contributors.

But this is hardly all o the spending in

which unions engaged to inuence the elec-

tion. Union political action committees

(PACs) made their own contributions, to the

tune o $66.4 million, to congressional and

presidential races12. These PACs contributed

another $6.75 million to national parties; 92

percent o these contributions went to Demo-

cratic candidates or to the Democratic Partyitsel 13.

And this does not cover all costs incurred

by unions to support candidates. Unions are

also allowed to advertise to their own mem-

bers, encouraging them to vote one way or

another. FEC flings are required to identiy

specifc unions’ independent expenditures.

These data are illuminating.

The SEIU spent over $6.5 million in 2008canvassing, producing yers, T-shirts, but-

tons and the like to persuade their members

to vote or Barack Obama. The American

Federation o State, County, and Municipal

Workers spent $12.5 million communicating

to their members about the election, although

much o the money went to supporting Hill-



S e p t e m b e r 2 0 0 9 2 7

demonstrates SEIU’s commitment to increas-

ing its fnancial weight in the political arena:

“By December o 2009, 20% o every Local

Union’s members should be giving at an aver-

age o at least $7 per month to COPE…Oncea Local has achieved this 20% goal, the Lo-

cal should increase the number o members

giving at an average o $7 or more per month

by at least 10% each year until a majority o

members are giving at this level…”17

The SEIU hopes that it can persuade its

members to donate an average o $16.80 a

year, and eventually achieve a rate o at least

$39 a year. That is, SEIU aspires to talk itsmembership into $78 million o political und-

ing each year18.

The SEIU appears to be the union most

explicitly demanding that its membership f-

nance its political agenda, by writing into its

constitution a requirement that local unions

und its political endeavors. But the money to

und union 527s may come rom local and na-

tional union contributions (union 527s almost

exclusively)19, and thus ultimately rom unionmembership dues. And i there is one certain-

ty looming over the horizon, it is that enact-

ment o the Employee Free Choice Act with

a card-check provision would give unions

the means to rapidly expand their fnancial

bases. SEIU has made the equation explicit:

every union member shall ante $6 a year or

political unding. Other unions have similar

expectations or their members to generatepolitical unds, money to strengthen unions’

political inuence in government at all levels.

Yet a more responsible path or unions would

be to make sure that their pension promises

could be ulflled beore turning to the politi-

cal agenda.

untary donations, but under the threat o a

penalty o 50 percent (or higher, i the Inter-

national Executive Board so decides), local

unions come under pressure to mobilize “vol-

untary” contributions o at least $6 a yearrom each member. These contributions can-

not be considered voluntary.

This policy, however, highlights the big-

gest problem with respect to union political

spending. Such spending, by law, is supposed

to come rom voluntary contributions, not

the dues that members are required to pay. I

unions divert dues money to political action,

union members may discover that their dueswere earmarked to support candidates they

oppose.

O course, union leaders can claim that

they know best which candidates, i elected,

will strengthen unions and beneft workers,

but that is a patronizing attitude. Unions

were not created because workers did not

know what policies were best or them. They

were created to unite workers into a collective

bargaining entity with more power than anyone worker could exercise. Any worker has

the right to be upset that union monies are

spent to support candidates that union lead-

ers decide are the best, rather than spent on

the worker’s preerred candidate, or, better,

on representational activities alone.

Federal law prohibits the use o compul-

sory union dues or political purposes. This

is why the SEIU encourages locals to increasevoluntary contributions to COPE, and why

the union constitution no longer states that

COPE unds may come rom local union dues

revenue.

The ollowing language in resolutions

made by SEIU during its 2008 convention




and retirees. In contrast, a separate und or

the union’s own employees, numbering 1,305,

participants was 91% unded. Even better, the

pension und or SEIU ofcers and employees,

which had 6,595 members, was 103% und-

ed.”

The SEIU lambasted the article and

claimed that the SEIU National Industry Pen-

sion Fund had achieved high unding levels,

92 percent in 2006, and 96 percent in 200821.

Now, perhaps the union’s internal calcula-tions showed the SEIU pension plan was in

good shape, but in 2009, the SEIU National

Pension Fund reported to the DOL that it was

in critical status—a sign o serious unding de-

fciencies that suggests the SEIU’s arguments

were ignorant at best, and disingenuous or

worse i they were aware o these problems22.

In addition, three o SEIU’s pension unds

were in endangered status as o 2008, and this

year the 1199 Pension Fund declared criticalstatus23.

The Local 32BJ District 36 Building Main-

tenance Pension and the Local 32BJ District

36 Building Operators Pensions cover togeth-

er 7,000 people. And the SEIU 1199 Greater

New York Pension covers another 29,000, or

36,000 in New York in all.

Under the 2006 Act, these workers are

not allowed to win increased pensions in newrounds o bargaining, and in many cases have

to accept lower beneft accrual rates—which

imply smaller annuities than promised and ex-

pected—until the und is ully unded again.

While these endangered plans may cause

hardship or some or many o the 36,000

workers when they retire, the status o the

The ollowing eight case histories illus-

trate issues related to union pension

plans. Until the passage o the Pension

Protection Act o 2006, the Department o

Labor did not have the authority to impose

corrective action on underunded plans. As a

result, a number o weaknesses in union pen-

sions grew to harm rank-and-fle union mem-

bers. There have been instances o outright

wrong-doing—embezzlement, kickbacks, and

similar illegal activities—but ar more harmhas been done by the implicit orms o negli-

gence discussed in previous sections.

Fortunately, some o the issues discussed

in these case histories were addressed in the

Pension Protection Act, to the consterna-

tion o union leadership. These case histo-

ries should act as cautionary tales that teach

union members to be vigilant, to use all the

tools the law gives them to keep a close eye on

their pensions, and their leaders.

Service Employees

International Union

On July 11, 2008, in response to an ar-

ticle published in the July 9 edition o the

New York Sun by Diana Furchtgott-Roth on

the state o union pensions

20

, the Service Em-ployees International Union (SEIU) issued a

blistering press release. The article stated “Yet

in 2006, the SEIU National Industry Pension

Plan, a plan or the rank-and-fle members,

covering 100,787 workers, was 75% unded.

That is, it had three-ourths o the money it

needed to pay beneft obligations o workers

XII. Eight Case Histories



S e p t e m b e r 2 0 0 9 2 9

1199 Pension Fund in 2007, when it still had

a strong unding ratio, but the national union

leaders that claim credit or keeping the na-

tional plan well-unded did not deem ft to

share this insight with their locals.The end result is stunning in its inequity:

1,000 District o Columbia and Maryland

nurses are acing a ailing pension while the

national pension—not to mention the 102

percent unded ofcer pension plan—our-

ishes. In the uture, beore bragging about

its stability, SEIU should ensure that all o its

workers are reaping the benefts.

E.I. DuPont

1199 Pension Fund (a distinct und rom the

Greater New York plan) endangers the uture

income security o close to 200,000 health

care workers in New York, Maryland, Mas-

sachusetts, and New Jersey. At the beginningo 2007, this und had assets o $8.3 billion,

but in the two years ollowing the last Form

5500 fled by the und, it lost one-third o its

value24. This would be enough to reduce the

unding ratio rom 90 percent to 60 percent,

assuming the one-third loss occurred starting

in 2008. It would be easy to blame this decline

on the economy and the stock market, and it

is true that many pensions lost a lot o moneyas a result o the recession and stock-market

slide. But it is also true that a large number o

well-unded pensions did not lose one-third o

their assets as a result o the recession.

While the SEIU National Industry Pension

Fund bragged about its well-unded plan, a

2007 letter rom the trustees reveals that the

stability was achieved at the cost o cutting

union members’ uture pensions25. The 1199

representatives, however, according to onesource, reused to bring a reduction in uture

pension benefts to the bargaining table when

they entered into negotiations to repair the

ailing plan26. Their desire to protect rank-

and-fle pensions would be admirable, were

they not holding out or a promise they can-

not keep—not without more money that nei-

ther the union nor employers have.

The national union is not responsible orthe unding status o a district pension und.

The leaders o that district, together with the

employers’ representatives, have that respon-

sibility. But the national pension und had

the right idea, cutting uture benefts in the

ace o an international economic crisis. This

sort o advice would have been critical to the

In 2007, the union pension plan or E.I.

Du Pont De Nemours & Co. covered 157,794

people, 31,693 o them current workers. This

plan was negotiated by the International

Brotherhood o DuPont Workers (IBDW),

which represents employees o E.I. Du Pont

and its subsidiaries. In 2007, the plan was109 percent unded, and according to pension

documents is expected to recover rom 2008

losses by the end o 2009.

It is not in any way the best-unded pen-

sion plan in the country, but the DuPont Pen-

sion Plan is well ahead o the curve among

plans o comparable size, and even perormed

better than average in 2008. It is at no risk o

alling into endangered or critical status, andDuPont and the International Brotherhood o

DuPont Workers entered arbitration in May

regarding the continuation o the defned ben-

eft pension plan.

That the DuPont Pension Plan is a single-

employer plan gives it natural advantages in

easing negotiations and ocusing workers’ at-




and $20.67 billion in 200729. Given that the

und’s liabilities were more than $44 billion,

orming a nearly $24 billion defcit in 2007,

it is unsurprising that the und declared itsel

to be in critical status when frst required in2008.

However, one o the Central States’ sis-

ter unds, the Western Conerence Teamsters

pension, was in no such difculty and antici-

pates ew, i any problems in the uture. In

2007, its unded ratio was 80 percent, and ac-

cording to the annual unding notice rom the

und’s actuary, it was 97 percent unded in

200830

. It has dropped back to an 85 percentunding level in 2009, but reasonably expects

to be ully unded without any extraordinary

eorts.

So, what is the dierence between these

unds?

The major dierences are three. First, the

Central States Fund is managed by J.P. Mor-

gan and Goldman Sachs, a holdover rom a

court decision made in response to the late

Jimmy Hoa’s mismanagement o Team-ster pension unds decades ago. (Hoa was

president o the international union in 1958-

1971.) The Western Conerence is controlled

by the union and participating employers. It

is tempting or unions and their supporters

to blame the banks or poor perormance,

except that the same problems have plagued

union pensions not controlled by court-ap-

pointed banks.Another dierence is investment strategy,

a ar more persuasive point. Over two-thirds

o the Central States investments are in stocks,

making its und value heavily dependent on

market perormance31. The Western Coner-

ence unds are slanted towards more conser-

vative, diversifed assets. In 2004, the West-

tention on the responsibilities o union and

employer alike to keep the pension well-und-

ed. But single-employer plans are not certain

to do well, even among smaller plans.

According to IBDW, their organizationis unique. All members o the union – rank-

and-fle, ofcers, and even the executive

board – are employees o DuPont or its sub-

sidiaries and spin-o companies27. There are

eight locals in IBDW, and the parent union’s

president can note how local presidents deal

with the year’s challenges individually. When

DuPont considers layos, President Jim Flick-

inger o IBDW is deending his own interestsas well as that o his constituents. And when

the pension plan alls behind on its unding,

every member o union management aces a

real threat to his or her retirement security.

This may not be the perect union para-

digm, but there is a great deal to be said or

the act that its leadership is personally invest-

ed in the well-being o its members.

International

Brotherhood of Teamsters

At the end o 2006, the Central States Pen-

sion Fund covered over 451,000 workers, with

about 155,000 o them current workers. Yet

on January 1, 2008, UPS, the parcel delivery

company, bought out the contracts or 44,000

o those workers, paying the und $6.1 billionand shrinking the number o current workers

covered by the plan to about 106,000. The

cash inusion, however, was not enough, and

the und determined it was in critical status

in 200828. At the end o the frst quarter o

2009, the und had $15.66 billion in assets—

down rom $17.36 billion at the end o 2008



S e p t e m b e r 2 0 0 9 3 1

maximum $12,870 a year or a worker with

30 years o service37. This was averted only

by the Central States and its contributing em-

ployers developing a rescue plan that involved

reducing uture accruals o benefts, shu-ing health and welare beneft contributions

to the pension plan, and requesting the IRS

to waive minimum unding standards or a

decade38. This waiver was contingent on the

Central States maintaining a stable unding

ratio, which they ailed to do, and the und

applied or a second waiver in 200839.

In 2003, the Western Conerence em-

barked on a similar recovery plan. Poor eco-nomic perormance had caused $5 billion in

losses that required the und to cut accru-

als or uture benefts40. They expected cur-

rent benefts to soon exceed the und’s assets,

which could hoist red ags at the IRS.

When the Western Conerence reduced

uture benefts to protect the plan’s unding

ratio, it was already ully unded41. The actu-

aries knew that uture benefts had to be re-

duced to keep the und’s ratio rom slippingin the uture, and did so to ensure that the

benefts the rank-and-fle had accrued would

be protected.

The Central States, meanwhile, was trying

to portray its und’s critical status as some-

how positive. It argued in a 2008 newsletter

to members that endangered plans are re-

quired to signifcantly increase contribution

rates, develop a strict 10-year recovery plan,and possibly eliminate uture beneft accru-

als42. The “red zone” – reserved or plans in

critical condition – allows a longer recovery

period and does not allow the und to reduce

beneft accruals below the Central States’

current rate. While Central States did not

lie about regulations, this rosy picture was a

ern Conerence trustees held 56 percent o the

und in government and corporate bonds, 39

percent in stock, and 5 percent in real estate32.

The Central States Fund’s latest flings list 1.5

percent o their assets held in “other, primar-ily real estate related” sources33. But this is

a disingenuous statistic, given that Central

States’ und holds 33 percent o its assets in

“fxed income” investments34. According to a

Charles Schwab tutorial, “fxed income” se-

curities include real estate investment trusts35

and GNMA mortgage-backed securities36. It

is thereore uncertain how much o the und is

and was invested in real estate, an ironic stateo aairs given that the current pension struc-

ture came about as a result o Central States’

historical real estate investments maintained

by organized crime.

Arguments have been made that the bank-

managed pension und has a bias towards

high-risk, high-reward investments, rather

than the more conservative, and thus more

stable, investments the Western Conerence

engages in. But saying that the trustees haveno capacity to suggest or require more con-

servative investment strategies is dubious. It

is ar more likely that the trustees have been

in avor o the high-risk strategy in the hope

o high rewards. Ater all, the pensions on the

line are or rank-and-fle members, not the

trustees.

The third dierence between the unds

is one that Central States can’t blame on theeconomy, the fduciary managers, or employ-

ers. In 2001-03, the Central States struggled

with poor perormance and a alling und-

ing percentage. There were serious concerns

that the Pension Beneft Guaranty Corpora-

tion (PBGC) would be orced to take over

the und, reducing guaranteed pensions to a




estate deal46. An increase in the unding ra-

tio two years ater replacement o most o the

board o trustees indicated that the und was

not doing as poorly as it might have.

However, at 55 percent, the und was not

doing well. Like many pension unds, ater the

stock market bull market o the late 1990s,

the Plumbers’ pension had a huge “amortized

credit”—an imaginary number supposedly

reecting how much excess money the und

had due to previous contributions above mini-

mum required payments and appreciation o

investments.

Every year, a pension plan accumulatesadditional pension costs due to the additional

year o service achieved by workers. This is

called the “normal cost”, and a pension plan

that ails to acquire at least as much money

over the year is alling behind. When a plan

alls behind in its unding due to poor market

perormance, skipped payments, or similar

events, it racks up debt that must be paid back

in annual payments, much like a mortgage.

When a plan is ahead on unding, it accumu-lates credits that it can use to reduce uture

minimum annual contributions.

The Plumbers’ board o trustees has, in

recent years, been paying its normal cost, but

has not been paying back its debt accumulat-

ed in previous years. They use accumulated

credits to reduce annual payments on the pen-

sion’s debt, reducing money owed on paper

while the plan still lags in value.The Pension Protection Act lessened the

Plumbers’ ability to use this fnancial sleight

o hand to keep rom shoring up its low und-

ing ratio, but the law came into eect only in

2008, delaying when we will see its eects on

the Plumbers. It is yet uncertain why EBSA

has not recognized the Plumbers as being in

sugar-coated —and sel-serving—way or the

Central States administrators to tell members

that their pension plan won’t be solvent or

more than a decade, while better-o plans

will have recovered aster, not to mention the

act that the plan would still have one-hal or

less o assets needed to pay all obligations.

The Teamsters or a Democratic Union43,

a dissident action within the union, recog-

nized that the administrators were not telling

them enough. This action thought Teamsters

members ought to be able to see what was

being done with their retirement money and

demand accountability or poor perormance.They asked to see the plan’s quarterly fnan-

cial reports, and when that ailed, sued or ac-

cess44.

The problem with the Central States has

been going on or years without stop. Only

the Pension Protection Act, fercely opposed

by the Teamsters, orced the Central States

to take decisive action to shore up its pension

und. The Western Conerence und shows

that it is possible or a plan to remain solvent

even in deteriorated economic conditions. All

it takes is a little oresight, a little restraint,

and a willingness to explain yoursel. The

Central States seems to have none o these.

Plumbers and Pipeftters

In 2008 the Plumbers and Pipeftters Na-

tional Pension Fund (also called the UA Na-

tional Pension) covered over 153,129 people.

In 2008, it had a unding ratio o 55 percent45.

This was a avorable and welcome develop-

ment or a union with problems. In 2004,

most o the board membership was ousted

ater a lengthy lawsuit over a poorly-run real



S e p t e m b e r 2 0 0 9 3 3

some unathomable reason, the administra-

tors agreed that i the property sold or more

than $44 million, Sardagna would receive 99

percent o the profts above that amount, even

though the union had put $34 million at risk.Furthermore, they agreed to sell the property

only with Sardagna’s consent49.

Had they known o it, this should have

been a warning bell to union members. Not

only was a man totally unconnected with the

union given power over deciding the ate o a

large asset o their pension und, but money

earned by union investments was going to be

lost to an outsider, with the consent o theirunion leaders. These acts alone suggested the

possibility o a corrupt relationship between

Sardagna and union ofcials.

The union itsel, had sta or ofcers made

a diligent inquiry, would have known that

Sardagna’s savings and loan had collapsed

on charges o bad loans and accusations o

raud (or which he was later acquitted). The

act that he blamed the collapse on bad loans

related to the property later bought by theunion pension und ought to have been an-

other warning sign. But under the advice o its

investment manager, William Seay, the pen-

sion trust went ahead with the purchase.

However, when the pension trust was

sued by the Caliornia government to clean

up the landfll, the trustees turned around

and sued the oil companies that they said

were responsible or contaminating it. Theyeventually discovered that their und adminis-

trator, William Seay, who had recommended

the purchase, had been a part-owner o the

land. Moreover, he contributed to unounded

union optimism about the value o the land,

encouraging the union to continue to invest

up to its eventual $34 million despite hety

critical condition, since their 2008 plan sum-

mary admits to a 55 percent unding ratio.

Southern California,Arizona, Colorado

and Nevada Glaziers

The pension trust o the South Carolina,

Arizona, Colorado & Southern Nevada Gla-

ziers, Architectural Metal & Glass Workers

Pension Plan is in poor condition. In 2007, it

had 12 percent o the assets necessary to sup-port present and uture retirees, or a defcit

o about $234 million. In 2008, the frst year

in which such a designation was required,

the pension trust fled “critical status” noti-

fcation with DOL47. The union had plenty

o scapegoats to blame or this poor peror-

mance—and did. Unortunately, the explana-

tions all reected poorly on the union’s lead-

ers’ judgment.

In 1989, two men named Robert Ferranteand Peter Sardagna convinced the Glaziers

pension managers to invest $34 million in

a parcel o land in Carson, Caliornia. The

land, which had previously been a landfll,

was to be cleaned and resold as a shopping

center, and so the pension und could expect

to realize a proft. The administrators contin-

ued to pay Sardagna as an independent con-

sultant until 1995

48

.The union and plan administrators then

agreed to the creation o a tangled fnancial re-

lationship: the Glaziers’ pension plan ormed

a limited-liability corporation to manage the

property and agreed that Sardagna would

aim to earn the pension a 15% return, plus

$6 million (about $10 million in total). For




poorly. Unions can easily blame und manag-

ers or poor perormance, but when the und

is ree to choose its own administrators, the

ault lies with the overseers – union leaders

appointed to the board o trustees.

United Auto Workers

The UAW is practically synonymous with

what were the abled Big Three o the Ameri-

can auto industry—General Motors, Ford,

and Chrysler. The trio was considered in the

second hal o the 20th century to be a cor-

nerstone o American manuacturing. That

history explains much o the 2008-09 panic

around the GM and Chrysler’s bankruptcies.

Contract negotiations between UAW and

the automobile companies have always been

tense, in part because o the energy UAW puts

into doing the best it can or its workers.

This history paints a rosy picture o the

union, and or the most part, it is deserved.

While UAW-negotiated wages have beencompetitive with that o Toyota’s U.S.-based

actories, hourly labor costs—wages and

benefts—have run close to $70 at union

plants, compared to about $46 at some non-

union plants, and even lower in the South54.

UAW’s negotiated pension plans were doing

well. In 2006, the UAW Chrysler plan, cov-

ering 52,500 active workers, was 87 percent

unded. The UAW GM plan, covering 80,000

active workers, was 102 percent unded. And

the UAW Ford plan, covering 85,200 active

workers, was 105 percent unded in 2005, the

last year available.

Part o this health came rom the UAW’s

wisely giving concessions to the companies

in 2003, when the Big Three were no longer

clean-up costs needed to make it a viable con-

struction site. I that were not bad enough, he

had once been a limited partner in the origi-

nal owner o the property, and had “long ties”

with Robert Ferrante50.Eventually, the union leaders wised up

and sued Sardagna or ull ownership o the

ormer landfll when he held up a possible sale

o the property to a man hoping to develop

it or a National Football League stadium.

This lengthy litigation ended in the und’s a-

vor, but the opportunity had passed, and the

land’s value is still uncertain as o mid-2009.

In 2005, the union won back $243,945rom kickbacks Ferrante accepted rom a real

estate developer51, and in 2003 Seay was con-

victed on charges o mail raud or deceiving

the union regarding his fnancial interest in

their investments52. The Glaziers fnally sold

the property in 2004 or $30 million, at a

loss, to a development group looking to create

a shopping center and residential area53. It is

possible the site will eventually show a proft,

but the rank-and-fle o the Glaziers will notrealize these gains.

Technically, the wrongdoers in this case

were Sardagna, Ferrante, and Seay. But the

pension trustees saw nothing wrong with their

investment advice and Sardagna’s question-

able fnancial arrangements until the corrup-

tion became ully evident. The trustees were

not skeptical and vigilant rom the beginning,

as their fduciary responsibility required. Theunion leadership gets points or ousting Seay,

and or disentangling itsel rom Sardagna.

But it is the union’s responsibility to appoint

responsible managers to its pension unds, and

to perorm due diligence in ensuring the man-

agers did their jobs. Given Seay’s history, it

appears he managed the other Glaziers’ unds



S e p t e m b e r 2 0 0 9 3 5

well-unded that union leaders lack. Even pri-

vately-owned companies are orced to answer

awkward questions i they let their pension

unding slip, while publicly-owned companies

could ace losses in market value i a ailing

pension becomes public knowledge. Unions,

however, have a host o scapegoats available

i it looks like their pensions are in trouble.

I companies have much to gain rom

keeping their union pension plans unded,

giving single- employer plans a unding edge,

why do unions preer to get companies to buy

into their multiemployer plans? On paper, a

multiemployer plan sounds like a good idea insome union industries, such as construction,

where workers may change jobs oten. But

that beneft is diminished i the pension plan

isn’t well-unded.

Some unions argue that giving control o

the unds over to the company is worse than

leaving it in the hands o trustees “who are

legally obligated to act in the best interests o

the participants.”58 This is a disingenuous ar-

gument that assumes the same principles o

unding and responsibility do not apply to

single-employer pension trustees. Given re-

cent history o union trusts using their unds

as leverage against corporations that don’t

adhere to their worldview59, one possibility

is that unions are loathe to lose control o

these large unds as tools or their agenda. O

course, that may not be in the union mem-

bers’ best interests.

Sheet Metal Workers

Starting in 2008, the Sheet Metal Work-

ers National Pension Fund began a desper-

ate attempt to rebuild its ailing pension

ush with profts. While UAW has resisted

company proposals to reduce uture pension

benefts, the union, recognizing the compa-

nies’ narrow circumstances, accepted 2004-

2008 contracts with no new costs55.But unlike many o the case studies pre-

sented here, the UAW pension unds are spon-

sored by private corporations. With pension

obligations a part o the auto companies’ bal-

ance sheets, they truly have a vested interest

in keeping the unds aoat. The Big Three

unds are doing well, i not equally so.

Like the Western Conerence o Team-

sters, the GM pension trustees reacted to thestock market uctuations early in the decade

by shiting to a more bond-heavy investment

strategy56. Ford suered more losses ater

2006 because it did not subscribe to GM’s

investment philosophy. Chrysler’s und ater

2006 is a mystery because the company was

bought out by a private equity group which

did not have the disclosure obligations o a

public company. None o these unds has even

reached endangered status.In act, in 2003, GM took it upon itsel to

sell $20 billion o bonds and contribute the

proceeds to its pension und to cover short-

alls, long beore the Pension Protection Act

would have orced the company to notiy its

workers o a possibly critical status57.

This highlights some o the strengths o

single-employer plans mentioned earlier in

this paper. Single-employer plans, like the BigThree’s plans, give employers the responsibil-

ity or maintaining the pension und, while

multiemployer plans have nebulous responsi-

bility, making it difcult to determine who,

i anyone, should make payments i the und

experiences shortalls. Furthermore, com-

panies have an incentive to keep their plans




they ailed to put equal eort into securing

those benefts. Because rank-and-fle workers

likely believed that benefts earned could not

be reduced, there was little incentive to devot-

ing energy to matching assets and liabilities—until orced by declining perormance met-

rics. And even then, the leadership dissembled

to conceal the extent o the problem.

In a 2007 letter to the rank-and-fle, chair-

man Michael Sullivan assured his constituents

that “critical” status “does NOT mean that

the NPF [National Pension Fund] is unable

to pay pensions. Rather, it means the actuary

projects that i the NPF did nothing, it would

have a unding defciency within the next ew

years.”64 This statement is grossly misleading.

The Pension Beneft Guaranty Corporation

has assumed control over pension plans better

unded than the Sheet Metal Workers, mean-

ing that to the government, a plan that is 50

percent unded is in danger o being unable to

meet its obligations.65 Furthermore, the claim

that “critical” status simply means the plan

will have a unding defciency in the utureglosses over the und’s severe actuarial def-

ciency at the time o writing.

As a result o its critical status, the und

has had to cut back on many o its “adjust-

able” benefts. The cost-o-living adjustment

was denied to new retirees and rolled back or

those receiving it. Early retirement benefts

were reduced. Lump sums and other optional

orms o payment were eliminated. And, de-

pending on the plan negotiated by individualemployers, either beneft accrual was reduced

to a minimum o 1 percent and certain early

retirement options disallowed, or beneft ac-

crual was to be set at 1.5 percent and annual

contributions increased by 7 percent66.

Union presidents do not like coming to

plan. Covering 136,000 workers, 70,000

o whom were still working, the plan had

slightly more than 50 percent o the assets

required to cover its liabilities60. To comply

with the Pension Protection Act, the SheetMetal Workers, whose pension und was in

critical status, had to develop a 10-year plan

to rebuild its unding ratio and ensure that it

could pay promised pensions.

Like many ailing pension plans, the Sheet

Metal Workers had put much o its assets in

stocks—62 percent in 2007. Naturally, this

let the union vulnerable to stock market

downturns, as its stock holdings lost 39 per-

cent o their value in 2008, according to pen-

sion fnancial documents61. A chart provided

by the union to discuss 2008 perormance

revealed another difculty—the increasing

costs o benefts. Through 2007, beneft costs

increased steadily, while assets lagged in 2003

to 200562. Perhaps more clearly than any other

union pension plan, the Sheet Metal Workers’

plan demonstrates one o the most common

pitalls to a collectively-bargained, union-runpension plan.

Beore 2008, the pension plan oered

a wide variety o extra options: an annual

cost-o-living adjustment to the pension63, a

lump-sum payment option, a subsidized early

retirement plan, a disability payment equal to

maximum possible early retirement benefts,

a 10-year guaranteed payment (that would go

to a designated benefciary ater the retiree’s

death), and others. The sum o these benefts(most especially the cost-o-living adjustment)

contributed to steady increases in the plan’s

required annual benefts payout.

Despite all the promised retirement ben-

efts won by the Sheet Metal Workers’ leaders

through dogged bargaining with employers,



S e p t e m b e r 2 0 0 9 3 7

Another service created by UNITE-HERE

is the UNITE Here Fund Administrators,

which in 2007, according to its IRS Form

990, “perormed administration work or six

health & welare unds and fve retirementunds.” UNITE Here Fund Administrators

in 2007 charged $33 million or its services,

when it was carrying an employee payroll o

$15 million.

UNITE-HERE should be concerned about

the continuing entanglements o these three

organizations and their inherent conicts o

interest. The frst worrisome sign is that 21

UNITE-HERE leaders, including presidentBruce Raynor and hospitality industry pres-

ident John Wilhelm, are reported as having

a controlling relationship to Amalgamated

Bank. All told, UNITE-HERE holds a 57 per-

cent interest in the bank, and its leadership

earned $428,600 in compensation rom the

bank, an average o $20,400 each.

The UNITE HERE Retirement Fund has

a number o stock holdings, including all o

the common stock o Alico Services Corpora-tion. Alico, in turn, owns UNITE Here Fund

Administrators and the Amalgamated Lie In-

surance Company. It is one thing or UNITE-

HERE to have ounded all o these organiza-

tions, and quite another or UNITE-HERE’s

administration to directly control all o them.

It is a direct conict o interest to have pen-

sion unds invested in related organizations.

And although the DOL determined the ar-rangement was legal, it is decidedly odd that

in 2005 UNITE-HERE sought to purchase

15 percent o Alico’s stock67.

Technically, UNITE-HERE does not di-

rectly control UNITE Here Fund Administra-

tors. Its president, secretary, and vice presidents

are Ronald Minikes, Mark Schwartz, Michael

their members with bad news. The excerpt

rom Michael Sullivan’s early notifcation

demonstrates that. That tendency towards

dissembling carries over into contract renew-

als, when union leaders highlight beneftswon without acknowledging the costs. An in-

creased pension is not simply a victory or a

union. It is an obligation and a trust on the

part o the pension sponsor, oten the union

that won the beneft. The eort put into main-

taining that pension needs to be signifcant,

lest the “adjustable” portions o that pension

be lost.

UNITE Here Fund

Administrators, Inc.

In 1922 the Amalgamated Clothing

Workers, a predecessor to the present-day

UNITE-HERE union, ounded a bank in

Chicago to service the fnancial needs o

blue-collar working people, especially union

members. (The ACW ounded a sister bank inNew York City in 1923.) The Amalgamated

Bank o Chicago’s board is today still heavily

weighted with union leaders and supporters,

and retains connections to its ounding orga-

nization. UNITE-HERE’s relationship with

the fnancial services industry has given it a

leg up in creating fnancial services entities to

serve the union world and the public.

Among these was Amalgamated Lie,ounded in 1943 to oer insurance services

to the UNITE-HERE (then Amalgamated

Clothing Workers o America) unds. While

originally created to serve members o the

UNITE-HERE constituent unions, in 1992,

Amalgamated Lie received permission to sell

its services to the public.




why rank-and-fle members are being charged

more than twice the amount the union’s sta

pays or plan administration.

Presumably these relationships have ex-

isted long enough that the ederal governmentwould have identifed any wrongdoing, but

the arrangement still produces a curious set

o fnancial relationships. There is no clear ex-

planation why the sta pension plan is man-

aged by a dierent entity than the rank-and-

fle pension, one that does its job more cheaply

and more eectively68. Non-proft records give

no indication how UNITE-HERE’s adminis-

trators keep separate the interests o their nu-merous fnancial holdings. And nothing sug-

gests how the leaders o UNITE-HERE fnd

time to manage their union and make signif-

cant bonuses rom Amalgamated Bank at the

same time.

The case o UNITE Here Fund Admin-

istrators highlights another problem in the

world o union fnances, especially defned

beneft unds. Even with the level o transpar-

ency required, administrators can easily cloakadministrative and management ees in mys-

tery to keep workers rom understanding just

how their pensions are being managed. And

in complex situations like that o UNITE-

HERE, administrators can generate oppor-

tunities or proft even without resorting to

embezzlement.

Hirsch, and Paul Mallen. However, all save

Ronald Minikes are vice presidents at Amal-

gamated Lie, all earning salaries in excess

or $100,000 or their services. But the direc-

tors o UNITE Here Fund Administrators in-clude such people as Bruce Raynor, Noel Bea-

sley, and Edgar Romney, among other senior

UNITE-HERE ofcers. Many o these same

people are the trustees o the UNITE-HERE

National Retirement Fund, which manages

pensions or the union’s own employees.

Despite the Retirement Fund’s assertion

that “the Fund has a written code o ethics

which covers conicts o interest,” it is hardto imagine that the competing interests o

und administrators, the union, the pension

und itsel, Amalgamated Lie, and the Amal-

gamated Bank have had no eect on UNITE-

HERE’s fnancial status. This is not to say the

und is in dire straits—in 2007, it was 83 per-

cent unded.

But Amalgamated Bank made close to

$2.8 million in 2007 rom investment ees

rom the UNITE HERE Retirement Fund,and the UNITE Here Fund Administrators

took $13 million in ees or managing the

und. The Fund Administrators’ ee amounts

to $47 per participant. In contrast, the sta

pension plan was administered by Alicare,

one o Amalgamated Lie’s companies, or

$129,000, $22 per participant. It is unclear



S e p t e m b e r 2 0 0 9 3 9

solvent. Furthermore, labor leaders ought to

recognize how benefts inuence overall pen-

sion costs and how this aects unding ratios.

Even modest sweetening o benefts can cause

large increases in uture obligations. In short,

beneft increases are not always in the work-

ers’ best interests. Unsustainable gains in ben-

efts lead to expectations that must ultimately

be disappointed.

Union leaders have ew incentives, how-ever, to take such a proactive, cautious view.

Their own pension plans are oten separate

rom their members’, meaning that a ailed

rank-and-fle pension plan has no personal,

pocketbook impact on the leaders. The sound

condition o union leaders’ pension plans rel-

ative to the condition o rank-and-fle plans

demonstrates that union leaders know how to

properly manage pension plans, and thereore

must have some idea about how they are en-dangering workers’ utures. Furthermore, the

act that ofcer pensions tend to be run by

ofcers themselves lends weight to the propo-

sition that individuals are ar better o when

in charge o their own fnancial uture.

The real problem is the opacity o union

pension fnancing, and the lack o account-

ability required o union leaders. The Pension

Protection Act has exposed more than 300poorly unded plans, orcing them to take ac-

count o their insecure positions and to make

eorts to rebuild. Yet, even with the Act,

bizarre entanglements like those o UNITE-

HERE are possible, and the expenses o some

pension unds are still shrouded in mystery.

The Act has had other positive eects,

Despite their rhetoric, unions cannot

deliver a retirement more secure than

that oered by non-union plans. By

standardized measurements, union-run pen-

sion plans are consistently worse than their

non-union counterparts. As noted above,

while 59 percent o union unds had 80 per-

cent or more o the assets needed to pay ex-

pected costs, 86 percent o non-union unds

did. Unions may promise their members supe-rior benefts, but they do not deliver.

Union leadership has contributed to this

problem by negotiating or pension benefts

that are more than aordable, oten seeking

pension increases in the ace o consistent

employer warnings about cost. Many unions

believe that increasing benefts is more impor-

tant than keeping pension unds ully unded.

Trustees need to make wise, inormed de-

cisions to help pension unds remain stable.Funds with conservative investment strate-

gies, especially those with lower stock and

real estate holdings, have ared ar better than

those with high-risk investment strategies.

But negotiators have to commit themselves

to supporting higher unding ratios, as well,

even though the riskier, more volatile growth

stocks chosen to pay or uture benefts are

also more volatile. The Pension ProtectionAct has gone a long way towards orcing

unions with troubled unds to negotiate or

more aordable benefts, but labor needs to

work more proactively. Ater all, it is cheap-

er to pay now to keep a und rom slipping

than to wait until assets depreciate and some-

one is required to pay again to keep the und

XIII. Conclusions




ensuring that rank-and-fle members will have

stable fnancial utures.

Unions are eager to point out the aws

in defned contribution plans, and certainly

the current economic downturn demon-

strates there are aws in these plans. But

defned beneft pension plans have ared just

as poorly, and unlike with a defned con-

tribution plan, workers cannot identiy the

exact loss and its impact on them. Given the

expertise union leaders have shown in man-

aging their own pensions, they could easily

oer advice to workers to help them protect

their utures.Whether workers continue to preer de-

fned beneft plans or push or more defned

contribution plans, it is clear they need to de-

mand more accountability rom union leader-

ship, to ensure that they can achieve the ben-

efts that union leaders promised them.

some lambasted by unions. Struggling plans

cannot increase benefts, and must limit non-

standard pension options, preventing plans

in poor condition rom urther expanding

pension costs. The Act also requires unions

to give participating employers options when

their workers’ pension plans are in trouble.

No union can simply demand an employer

increase contributions to support a agging

pension und.

Yet the law deals only with the symp-

toms o the problem. Forcing pension plans

to bring their accounting back into order is

a good idea, but ails to address the age-oldprincipal-agent problem. That is, union mem-

bers have ew assurances that the trustees o

their pension unds are truly acting in their

best interest. But the continued poor status o

union-run pension plans suggests that union

trustees are not adequately working towards



Appendices




annually. I she has $1,000,000 on the day sheretires, she will have more money than she needsbecause o the 3 percent interest rates.

So she looks at it another way. Say she wereretiring next year, and wanted her bank ac-count to have $50,000 in it. She would need to

invest $48,543.69, because (1.03) x $48,543.69= $50,000. I she wanted her bank accountto have $50,000 in it two years rom now, shewould have to invest $47,129.80 today, because(1.03)2 x $47,129.80 = $50,000.

The bank advises Susan that in its 3 percentretirement plan she must save $12,500 a year, orabout $1,050 a month.

I defned beneft pension plans were this sim-ple, companies would not have so much difcultywith them.

Unortunately, there are a number o compli-cations that make the people who plan and trackthese unds (called actuaries) have much morecalculating to do. Most o this work arises be-cause companies preer to make level payments

every year to stabilize cash ows. There are alsouncertainties that will arise over the years as theemployer’s payroll expands or shrinks, as com-pensation changes and as investment outcomesdeviate rom the original assumptions.

For all o these reasons, a plan may fnd itsel underunded or overunded.

Earnings

Most defned beneft pension und beneftsare a percentage o an employee’s annual earn-

ings. The ormula can be based on “average”earnings, as is Social Security, or it can be basedon the worker’s earnings during his last year orseveral years o work. In either case, it is tremen-dously difcult to predict how much a worker

will earn during his last several years, perhaps 20

Adefned beneft pension plan is a promiseto pay each participant a specifed sumo money during each year o retirement.

Such annual payments (usually divided into 12monthly disbursements) are called an annuity.

Annuities can result rom pension payments

and they can be purchased rom insurance com-panies. Let’s examine how this aspect o fnanceworks. The calculations that ollow are a simpli-fed model o how these plans work in the realworld. Few individuals make these calculations

themselves, but the example is intended to illus-trate a simplifed version o the process that pen-sion actuaries go through.

Imagine a 35-year-old woman (we will callher Susan) who wants an annuity to pay her$50,000 a year ater she retires69. I her bankpromises a at, unchanged interest rate o three

percent, she can plan her retirement income witha air degree o certainty. Susan will frst calculatehow much money she must have earning 3 per-cent to generate $50,000 a year ater she reaches

65. Then Susan will calculate how much a year(or a month) she must pay under her contract to

und the annuity. The younger she is when sheenters into the annuity contract, the more pay-ments she will make, but the lower each annualcontribution must be.

According to the Social Security Administra-

tion, a 35-year-old woman can expect to live ap-proximately 46.22 years, to age 81. Susan wantsto play it sae, so decides to assume that she willlive until age 85. Thus she will need the annuityto provide her payments or 20 years, rom age

65.Susan frst thinks that her annuity account

will need to have $1,000,000 ($50,000 x [85—65]) in it on the day she retires at age 65. Shequickly realizes that this is incorrect. Every year,

the remaining balance in her account will growby three percent, even as she withdraws $50,000

A p p e n d i x I :

How Do Pensions Work?



S e p t e m b e r 2 0 0 9 4 3

sion payments than workers who die at 70. I lie

expectancy increases over time by one year, the

und’s pension liability increases.

Real Fund Growth

Susan, above, could predict exactly how much

money she needed because she knew that the con-

tractual interest rate was three percent, and al-

ways would be. Defned beneft plans do not put

their money in banks, but invest in a wide vari-

ety o stocks, bonds, and other fnancial products

that give dierent returns on the investment. The

plans must assume that their money will grow at

some rate, but have no guarantee that it will. Thiscreates uncertainty about the uture value o the

und. As some o the experience cited in this pa-

per demonstrates, a defned beneft und may fnd

that it is overunded and can relax contributions,

or that it is underunded and needs to augment

contributions (assuming unchanged promised

pension benefts). In either situation, employers

and unions may have conicting ideas about what

should be done.

Normal Cost

The phrase “normal cost” reers to the annual

expected increase in pension obligations. Nor-

mal cost involves increases based on additional

experience earned by workers, increases due to

increased wages earned, and costs incurred by

adding workers into the pension und. Given the

model’s assumptions, normal cost is easy to cal-

culate. It is the minimum contribution one canmake to a und perorming within expectations

to ensure that the und will have all necessary as-

sets to pay uture obligations. As a result, it is a

good base or determining whether a und is re-

maining stable or slipping in its unding ratio.

years into the uture. So, or many workers, actu-

aries do not know in advance precisely how much

their accrued benefts will be. Actuaries address

this problem by making assumptions about the

percentage growth o an employee’s wages eachyear, and use that number in their calculations.

Experience

It is a common eature or one’s defned beneft

stream to be equal to a percentage o income per

year worked. For example, a retiree whose beneft

is one percent per year and who had worked or

the employer or 10 years would receive 10 percent

o his income, and i he had worked or 40 years,40 percent o his income. This requires actuar-

ies to develop models suggesting how much ex-

perience workers, on average, will have acquired

beore retiring or leaving the company. In other

words, individuals who work or the company or

10 years will earn a much smaller pension than

people who work there or 40 years.

Retirement Age

and Lifespan

I employees are promised the same pension

regardless o when they retire, then a person who

retires earlier costs the company more than one

who retires later. Consequently, pension plans re-

duce benefts or those who retire early and aug-

ment them or those who work beyond normal

retirement.

Companies keep track o traditional retire-

ment patterns, and try to estimate when their

workers will retire. I workers tend to retire earli-

er, companies will try to increase contributions so

that the employees can make larger payments or

a shorter period. Their estimates are important

or determining how much the company expects

to owe each year.

Liespan is another key variable. Workers

who live to be 90 will receive much more in pen-




plan that unction much like mortgages. Each

year, the amortized charges require a certain fat

payment be made to reduce it. A credit enters the

equation when changes in returns or assumptions

exceed expectations. Until recently, credits couldbe used to reduce the payments needed to be made

to cover amortized charges or even normal costs,

enabling a pension sponsor to avoid making any

payments to the und in years when credits were

high enough.

Actuarial Cost

The phrase “actuarial cost” reers to changes

in pension obligations caused by changes in the

underlying assumptions o the actuarial model.

These include changes in expected retirement age,

liespan, and returns on investment. As actuarial

cost oten involves unexpected changes in obli-

gations, it enters pension calculations as a series

o amortized charges—debts held by the pension



S e p t e m b e r 2 0 0 9 4 5

poorly-perorming unds up to their requiredunding ratio. Both choices reduce credits on the

plan’s books, and thereore increase assets allow-

able or calculating unding percentage.

Further, the Act requires at-risk plans (those

with less than 60 percent unding) to contribute

at least the normal cost each year.

The Pension Protection Act requires multiem-

ployer plans, beginning in 2008, to absorb their

entire unding liability as a 15-year debt. Mul-

tiemployer pension plans with less than 65 per-

cent unding will have to identiy themselves asin “critical” status, requiring the submission o a

plan to the Department o Labor to regain proper

unding status. Plans less than 80 percent unded

are “endangered” and required to adopt a plan

to revise benefts downward. The critical and en-

dangered provisions apply only to multiemployer

plans.

Both programs ollow a similar path: the

sponsor must provide two schedules to all partici-

pating parties. One reduces uture beneft accrual

(not currently-earned benefts), and the other in-

creases contributions. Plans in “critical” status

are specifcally required to develop more complex

action plans so that they can emerge rom that

status within 10 years.

In short, the Act’s central purpose is to re-

quire plans to remain well-unded; it lays down

specifc statutory provisions to prevent lapses.

Starting in 2008, critically-unded union

plans (assets are less than 65 percent o accrued

liabilities) came under stricter scrutiny, as unions

were orced to help employers and workers nego-

tiate a middle ground between decreased benefts

and increased costs.

Congress passed the Pension ProtectionAct o 2006 in response to growing wor-

ries over the defned beneft pensions o

workers across the country. It intended to require

pension sponsors to keep their unds actuarially

sound, in the hope that the statutory requirements

would ensure that no company or union would

make promises to workers they could not keep.

The law prohibits plans that are less than 80

percent unded rom increasing benefts. Further,

it prohibits unds with less than 60 percent o

their expected benefts rom paying benefciariesmore than a monthly annuity payment, that is, no

bonus payments.

Another o the Pension Protection Act’s main

provisions is a requirement that plans keep their

unds fnancially sound, or “on target,” with spe-

cifc provisions to prevent unding lapse. In 2008,

this target was 92 percent o all accrued liabili-

ties; it increases annually to 100 percent in 2011.

The act calls upon pension sponsors to con-

sider their total assets to be nominal assets—the

bonds, equities and real estate the plan owns—re-

duced by their credits or appreciation o securi-

ties that are publicly traded. These modifed as-

sets would be the test o how well unded a plan

is. Pension plans that are less than 80 percent

unded or 70 percent unded using “at-risk” as-

sumptions that mark up liabilities, are orbidden

rom reducing their calculated annual obligation

to pay benefts by applying credits. The sponsor

can otherwise use credits to reduce contributions

to the plan.

Sponsors can discard credits in order to in-

crease the value o their adjusted assets. This

allows them to use past overpayments to bring

A p p e n d i x I I :

The Pension Protection Act




percentage or dollar amount o contributions…to und all anticipated benefts, whether earned

or not earned…”71 This is called the “accrued li-

ability”. Another method, the “RPA 94 current

liability” is a standardized measure o benefts

earned to date, using ormalized interest rates.

The accrued liability is calculated with interest

rates chosen by plan administrators; the resulting

value varies with demographics and investment

policy, among other actors.

When comparing the unded ratios between

plans, the RPA 94 current liability is more ap-propriate that the accrued liability because it is

more conservative. The RPA 94 current liability

tends to be larger than the accrued liability, and

thereore leads to lower unded ratios than would

be achieved by using the accrued liability to de-

termine plan beneft costs. This bias, however,

aects all plans, whether union or non-union,

and does not aect the quality o comparisons

between plans. And the Vice President o the Pen-

sion Practice Council o the American Academy

o Actuaries, in a 2006 letter to Moody’s, opinedthat the RPA 94 current liability was likely an

appropriate measure to use in determining un-

derunding o pension plans, bringing into ques-

tion whether this tendency o the measurements

should be considered biased72.

It is difcult to determine a single way to ac-

curately measure assets and liabilities or defned

beneft pension plans, given the great uncertainty

inherent in these values. We have presented the

reasoning behind our choice o variables, and

some o the eects they have on our data.

The Labor Department’s Form 5500 hasseveral options or listing plan assets and

liabilities. How assets and liabilities are

presented can inuence the totals that matter or

determining whether a plan is adequately unded.

There are chiey two ways to calculate the as-

sets o a plan. The frst is to list the market value

o assets at the date a balance sheet or report is

issued. This is called the “current value” o assets.

The other starts with the und’s assumed growth

rate and averages any dierence between current

values and assumed growth over a period o time(usually fve years). This is called the “actuarial

value o assets”, or AVA70.

In this paper, the actuarial value o assets is

used to represent plan assets. Especially in a time

o great fnancial volatility, the asset smoothing

aspect o the AVA prevents a single uncharacteris-

tic year rom signifcantly altering a plan’s status.

As a result, a unded ratio presented in this paper

may show a traditionally well-unded plan to be

well o, even i it has suered in the recession. It

may also show a poorly unded plan to remain inthat status, even i it has experienced unexpected

growth in one year.

One signifcant eect o this methodology

would be that some plans may still be reecting

losses incurred as early as 2001, meaning that

some would not reect gains made in 2006 due

to still-unrealized losses in previous years. (“Un-

realized losses” means that an asset has been re-

tained, not sold, and has a market value below

acquisition cost.)

There are, in general, several ways to calcu-late plan liabilities. The frst is a usually a “level

A p p e n d i x I I I :

Data



S e p t e m b e r 2 0 0 9 4 7

1199 SEIU (27 April, 2009). “League and 1199 SEIU to Begin Negotiations Seeking Modifcations to the Col-

lective Bargaining Agreement.” Retrieved 18 July, 2009, rom: http://www.1199seiu.org/media/press.cm?pr_id=1420

Abrahamson, Alan (11 November, 1998). “Catalyst or Carson Stadium Deal Has Tangled History.” Los Ange-

les Times. Retrieved 1 July, 2009, rom: http://articles.latimes.com/1998/nov/11/news/mn-41577

Becker, Richard (January 2005). “Robbing Workers to Pay the Banks.” Socialism &Liberation Magazine On-

line. Retrieved 29 June, 2009, rom: http://socialismandliberation.org/mag/index.php?aid=140

Board o Trustees o the National Industry Pension Fund (12 October, 2007). “Reduction in Pension Plan For-

mula Eective January 1, 2008.” Retrieved 18 July, 2009, rom: http://www.workorceexchange.net/Portals/15/

announcements/SEIUPensionletter101207.pd

Board o Trustees, Plumbers and Pipeftters National Pension Fund (May 2009). 2008 Summary Annual Report.

Retrieved 5 August, 2009, rom: http://www.ppnp.org/docs/SAR_2008.pd

Bolen, J. Thomas. (3 April, 2009). “Actuarial Certifcation or 2009 as Required By the Pension Protection Act

o 2006.” Retrieved 8 July, 2009, rom: http://www.wctpension.org/whats_new/WCTPT_2009_PPA_Funding_

Certifcation_&_Explanation.pd

Bronstad, Amanda (2 June, 2003). “Tainted Carson Site Clears Hurdle Days Beore its NFL Nod.” Los Angeles

Business Journal . Retrieved 1 July, 2009, rom: http://www.thereelibrary.com/Tainted+Carson+site+clears+hur

dle+days+beore+its+NFL+nod-a0104259788

Center or Responsive Politics (2009). “Business-Labor-Ideology Split in PAC & Individual Donations to Candi-

dates and Parties.” OpenSecrets.org. Retrieved 14 July, 2009, rom: http://www.opensecrets.org/overview/blio.

php

Center or Responsive Politics (2009). “Federally Focused 527s by Industry.” OpenSecrets.org. Retrieved 14 July,

2009, rom: http://www.opensecrets.org/527s/527cmtes.php?level=I

Center or Responsive Politics (2009). “Pac Contributions to Federal Candidates.” OpenSecrets.org. Retrieved 14

July, 2009, rom: http://www.opensecrets.org/pacs/sector.php?cycle=2008&txt=P01

Central States Funds (31 March, 2009). Financial and Analytical Inormation. Retrieved 29 June, 2009, rom:

http://tdu.org/fles/CentralSFR2009.pd

Central States, Southeast and Southwest Areas Health and Welare and Pension Funds (Winter 2008). Team-

work. Retrieved 1 July, 2009, rom: https://www.centralstatesunds.org/CSF/Pd/teamworkWinter2008.pd

Central States, Southeast and Southwest Areas Health and Welare and Pension Funds (Spring 2008). Teamwork.

Retrieved 1 July, 2009, rom: https://www.centralstatesunds.org/CSF/Pd/teamworkSpring2008.pd

Central States, Southeast and Southwest Areas Health and Welare and Pension Funds (Fall 2003). Teamwork.

Retrieved 1 July, 2009, rom: https://www.centralstatesunds.org/CSF/Pd/teamworkFall2003new.pd

Christensen, Kevin (17 October, 2003). “International Union Oers Deal to Debt-Ridden Glaziers’ Pension

Plan.” San Diego Source

Department o Justice (4 May, 2005). “Former Pension Fund Consultant Sentenced.” Release No. 05-069. Re-

trieved 1 July, 2009, rom: http://www.usdoj.gov/usao/cac/pressroom/pr2005/06

Bibliography




Department o Labor (12 August, 2005). Prohibited Transaction Exemption 2005-11; Exemption Application

No. D-11185, et al. Federal Register (Vol 70, No 155)

Department o Labor Employee Beneft Security Administration (16 July, 2009). Critical and Endangered Status

Notices. Retrieved 16 July, 2009, rom: http://www.dol.gov/ebsa/criticalstatusnotices.html

Detroit Free Press Sta (16 October, 2003). “GM Executives Say UAW Pact Saves it Billions.” Detroit Free Press

Dolan, Peter B. (15 October, 2002). Department o Labor amicus brie fled in Southern Caliornia, et al., Gla-

ziers, et al., Pension Trust et al. v. William L Seay, et al., Peter A. Sardagna, et al. Retrieved 1 July, 2009, rom:

http://www.dol.gov/sol/media/bries/sardagna.htm

Fixed Income Investment Types. Retrieved 29 June, 2009, rom Charles Schwab web site: http://www.schwab.

com/public/schwab/investment_products/bonds_ixed_income/types?cmsid=P-982799&lvl1=investment_

products&lvl2=bonds_fxed_income

Furchtgott-Roth, Diana (Summer 2008). Union vs. Private Pension Plans: How Secure Are Union Members’

Retirements? Washington, D.C.: Hudson Institute

Georgeson (2009). “Annual Corporate Governance Review.” Retrieved 6 August, 2009, rom: http://www.

georgeson.com/usa/download/acgr/acgr2008.pd

International Brotherhood o DuPont Workers. “About the IBDW.” Retrieved 8 July, 2009, rom: http://www.

dupontworkers.com/about_the_ibdw.htm

Laborers’ International Union o North America National (Industrial) Pension Fund (10 December, 2007). Im-

portant Program Beneft Changes as o January 1, 2008. Retrieved 9 July, 2009, rom: http://www.lnip.org/

beneft_changes.htm

Laborers’ International Union o North America Board o Trustees National (Industrial) Pension Fund (21 April,

2008). Ofcial Notice o Endangered Status. Retrieved 9 July, 2009, rom: http://www.dol.gov/ebsa/pd/e-no-

tice0720080030.pd

Massey, Daniel (22 June, 2009). “Hospitals, Union Extend Mired Contract Talks.” Crain’s New York Business.

Retrieved 18 July, 2009, rom: http://www.crainsnewyork.com/article/20090622/FREE/906229979

Maynard, Micheline (30 November, 2008). “Autoworkers Worry More Givebacks Are in Store.” New York

Times. Retrieved 10 July, 2009, rom: http://www.nytimes.com/2008/12/01/business/01uaw.html

Mc Garr, Frank J. (1 June, 2009). Quarterly Report o Independent Special Counsel, Chao v. Estate o Frank

E. Fitzsimmons, et al., No. 78 C 342 (N.D. Ill., E.D.); Chao v. Robbins, et al., No. 78 C 4075 (N.D. Ill., E.D.);

and Chao v. Dorman, et al., No. 82 C 7951 (N.D. Ill., E.D.). Retrieved 26 June, 2009, rom: http://tdu.org/fles/

CentralStatesSpecialCounselReportJune2009.pd

McNamara, James (July-August 2005). “Top Plumbers International Ofcers Expelled.” Union Democracy

Review. Retrieved 9 July, 2009, rom: http://www.uniondemocracy.org/UDR/98-Top%20Plumbers%20Of-

cers%20expelled.htm

Moskalenko, Tanya (25 April, 2008). Notice o Critical Status or the Southern Caliornia, Arizona, Colorado

and Southern Nevada Glaziers, Architectural Metal and Glass Workers Pension Trust. Retrieved 8 July, 2009,

rom: http://www.dol.gov/ebsa/pd/c-notice0520080001a.pd

Pension Beneft Guaranty Corporation (30 June, 2009). PBGC Assumes Pension Plans at Precision Custom Com-

ponents LLC. Retrieved 13 July, 20009, rom: http://www.pbgc.gov/media/news-archive/news-releases/2009/

pr09-43.html

Principles or Responsible Investment (11 May, 2009). Letter to Senator Tom Harkin and Representative George



S e p t e m b e r 2 0 0 9 4 9

Miller. Retrieved 17 August 2009 , rom: http://www.acio.org/joinaunion/voiceatwork/eca/upload/Investor_

EFCA_Congressional_Letter.pd

Segal, Donald J. (March 2006). Comments on Proposed Rating Methodology or Companies Participating in

Multiemployer Defned Beneft Pension Plans. Retrieved 8 June, 2009, rom: http://www.actuary.org/pd/pen-

sion/moodys_march06.pd

Service Employees International Union (11 July, 2008). “Private Equity Bailout Industry Gets Its Facts Wrong.”

Retrieved 18 July, 2009, rom: http://www.commondreams.org/news2008/0711-09.htm

Service Employees International Union (2008a). Constitutional Amendments and Resolutions Adopted at the

2008 SEIU Convention. Retrieved 14 July, 2009, rom: http://www.seiu.org/images/pds/Constitutional_Amend-

ments_and_Resolutions_Adopted_at_the_2008_SEIU_Convention.pd

Service Employees International Union (2008b). SEIU 2008 Constitution and Bylaws as Adopted at the SEIU

2008 Convention

Shearin, Randall (2007). “Mixed-Use Along the 405.” Western Real Estate Business. Retrieved 1 July, 2009,

rom: http://westernrebusiness.com/articles/MAY07/eature11.html

Sheet Metal Workers’ National Pension Fund (19 May, 2009). “Revising the Rehabilitation Plan Webinar.” Pow-

erpoint Presentation. Retrieved 12 July, 2009, rom: http://www.smwnp.org/pds/NPFWebinar.pd

Sheet Metal Workers’ national pension Fund (24 April, 2009). Annual Funding Notice. Last retrieved 21 July,

2009, rom: http://www.smwnp.org/pds/2008AnnualFundingNotice.pd

Sheet Metal Workers’ National Pension Fund and its Subsidiaries (December 2008). Index of the Consolidated

Financial Statements for the Twelve Months Ended December 31, 2008 and 2007 . Retrieved 12 July, 2009,

rom: http://www.smwnp.org/pds/NPFInternal200812.pd

Sullivan, Michael J. (October 2007). A Message From Chairman Michael J. Sullivan About Upcoming Changes

to the National Pension Fund (NPF). Retrieved 13 July, 2009, rom: http://www.smwnp.org/pds/Chairmans-

Message.pd

Teamsters or a Democratic Union (21 August, 2008). “Central States Pension Fund Loses $3 Billion.” Retrieved

8 July, 2009, rom: http://tdu.org/node/2299

Teamsters or a Democratic Union (TDU) (18 October, 2006). “The Pension Protection Act: The Good, the Bad,

and the Ugly.” Retrieved 8 July, 2009, rom: http://www.tdu.org/node/384

Teamsters or a Democratic Union (TDU) (28 January, 2005). “Multiemployer Pension Funds Need Relie, Yes,

But First Union Reorm.” Retrieved 10 July, 2009, rom: http://www.tdu.org/node/837

Wall Street Journal Editorial Sta (28 July, 2008). Docking Paychecks or Politics. Wall Street Journal , p A14

Walsh, Marry William (25 November, 2008). “G.M.’s Pension Fund Stays Aoat, Against the Odds.” AARP

Bulletin Today. Retrieved 10 July, 2009, rom: http://bulletin.aarp.org/states/ny/articles/gms_pension_und_

stays_aoat_against_the_odds.html

Western Conerence o Teamsters Pension Plan. (Spring 2003). “What is Happening & Why.” Pension West ,

page 2. Retrieved 8 July, 2009, rom: http://www.wctpension.org/whats_new/pwnl_spring03/PWspr03.pd




1 “Tat-Hartley” reers to the Labor-Management Relations Act o 1947.

2 DOL EBSA, 2009. This represents only reported endangered and critical statuses. Backlogs in fling and

processing may have prevented certain unds rom being listed.

3 A number o types o plans were excluded rom the study. Frozen plans (those taking no new workers),

plans acquired by the Pension Beneft Guaranty Corporation (those in severe difculty or owned by deunct

companies), cash balance plans (a fxed-interest defned contribution plan), and plans in their frst or last

year o operation were excluded.

4 All reerences to unding use a ratio o the actuarial value o assets held by a given defned beneft pension

plan and the RPA ’94 current liability. For urther explanation o the use o these statistics, see the Appen-

dix.

5 A much larger number o plans are less than 65 percent unded than are in “critical” status. This is due inpart to the act that “critical” notifcations are required only by multiemployer pension plans, and due in

part to delays in status notifcation. Some plans in apparent critical status in 2006 may have improved in

the interim. Still, we will use “critical” to reer to all plans with unding ratios o 65 percent or less, or

consistency.

6 Teamsters or a Democratic Union (TDU), 2006

7 These data reer to plans either fling as a small plan or with 100 or ewer active participants. Frozen plans,

PBGC-run, and cash balance plans were omitted rom the study. Again, those in the frst or last year o

operation were omitted, as were plans covering a single individual.

8 AFSCME and APWU pensions are typically public pensions. Their accounting rules and regulations dier

rom those o private pensions, and so are beyond the scope o this paper.

9 The IBEW Pension Beneft Fund is a “non-qualifed” pension plan. It is a promise o a specifed rate ($4.50

per month per year o service) o payment. Because it lacks certain qualities, the und’s investment earnings

are taxed normally, and the und need not adhere to certain reporting and minimum contribution require-

ments. The most meaningul result o this is that the unding level o this und is not disclosed. However,

or completeness’ sake, we will report that in FY 2006, the plan paid out $101.7 million to benefciaries and

$46 million in contributions rom member dues, and $286.7 million in earnings. The und made $220.8

million over the year, a 12.6 percent gain.

10 Retrieved rom AFL-CIO’s EFCA page, http://www.acio.org/joinaunion/voiceatwork/eca/upload/Inves-

tor_EFCA_Congressional_Letter.pd

11 Center or Responsive Politics, Federally Focused 527s by Industry

12 Center or Responsive Politics, PACs13 Center or Responsive Politics, Business-Labor-Ideology Split

14 This includes the $27 million spent by the union and the $48 million the PAC fled as contributing, both

drawn rom FEC flings.

15 Service Employees International Union, 2008b, Article XV, Sec. 18(a)

16 Wall Street Journal , 2008

17 Service Employees International Union, 2008a, p 38

Endnotes



S e p t e m b e r 2 0 0 9 5 1

18 Based on calculations using the above commitment statement, and SEIU “ast acts”, http://www.seiu.org/a/

ourunion/ast-acts.php

19 While PAC contributions are oten rom individuals, FEC data show that PACs sometimes receive contribu-

tions rom unions. And unions qualiy as 527 groups, allowing them to contribute union unds or political

purposes.

20 Last retrieved 9 August, 2009, rom: http://www.nysun.com/opinion/holding-up-a-mirror-to-the-

seiu/81472/

21 Service Employees International Union, 2008

22 Retrieved rom EBSA’s Critical and Endangered Status notifcation page, http://www.dol.gov/ebsa/critical-

statusnotices.html

23 1199 SEIU, 2009

24 Massey, 2009

25 Board o Trustees o the National Industry Pension Fund, 2007

26 Massey, 2009

27 IBDW, 2009

28 Central States Funds, 2009, pg 3

29 Central States Funds, 2009, pg 8

30 Bolen, 2009, pg 1

31 Mc Garr, 2009, pg 8

32 32 Western Conerence o Teamsters Pension Plan, 2003, pg 2

33 33 Central States Funds, 2009, pg 9

34 34 Central States Funds, 2009, pg 9

35 Real estate investment trusts (REITs) are companies that own and lease or rent real estate property.

36 Charles Schwab, 2009

37 This is based on PBGC documents regarding multiemployer plans. The plan covers up to $11 per month per

year o service, and 75 percent o accrual rates above $11 per month per year, up to the maximum o $35.75

per month per year, or an annual pension o $429 per year o service. The explanation can be ound here:

http://www.pbgc.gov/practitioners/multiemployer-plans/content/page13111.html

38 Central States, Southeast and Southwest Areas Health and Welare and Pension Funds, 2003, pg 6-10

39 Central States, Southeast and Southwest Areas Health and Welare and Pension Funds, 2008b, pg 12

40 Western Conerence o Teamsters Pension Plan, 2003, pg 2

41 Western Conerence o Teamsters Pension Plan, 2003, pg 2

42 Central States, Southeast and Southwest Areas Health and Welare and Pension Funds, 2008a, pg 543 A rank-and-fle Teamsters organization dedicated to improving benefts and holding Teamster leadership

accountable or their actions

44 TDU, 2008

45 Board o Trustees, Plumbers and Pipeftters National Pension Fund, 2009, p 8

46 McNamara, 2005

47 Moskalenko, 2008




48 Bronstad, 2003

49 Dolan, 2002

50 Abrahamson, 1998

51 Department o Justice, 2005

52 Christensen, 2003

53 Shearin, 2007

54 Maynard, 2008

55 Detroit Free Press, 2003

56 Walsh, 2008

57 Walsh, 2008

58 TDU, 2005

59 The Georgeson report on corporate governance (see Reerences) shows that in 2008, the IBEW sponsored

a resolution or electing board directors by a majority vote at Washington Mutual. LIUNA sought to sepa-

rate the Chairman and CEO o Walgreen, and to link pay to perormance or Wal-Mart. SEIU sought to“redefne” what an independent Board Chairman is at Bank o America, and sought to separate the Board

Chairman and CEO o Wells Fargo. And the Teamsters sought to link pay to perormance at Allergan (a

health care company). Few o these companies have large union presences, nor are the unions even in the

same industry o the companies whose governance they wish to alter. And all o these proposals ollow a

union narrative o “good corporate governance”, rather than one o improved returns.

60 Sheet Metal Workers’ National Pension Fund, 2009, p 1

61 Sheet Metal Workers’ National Pension Fund and its Subsidiaries, 2008, p 2


63 Benefciaries received a thirteenth payment rom the pension und each year, equal to the percent increase

in cost o living.

64 Sullivan, 2007, p 1

65 Pension Beneft Guaranty Corporation, 2009, indicates assumption o the Precision Custom Components,

LLC pension plans, which were 58 percent and 53 percent unded.


67 Department o Labor, 2005, p 47236

68 “Eectively” indicates that o 2007, the sta plan was ully unded, while the rank-and-fle plan was not.

69 This assumes she intends to leave no money to heirs or other groups, wishing to consume all o her capital

over her lie.

70 I assumed growth is 8 percent, and in one year, growth is 13 percent, assets will be increased by 9 percent

each year on the actuarial value o assets.

71 Segal, 2006, p 2

72 Segal, 2006, p 2



1015 15th Street, NW, Sixth FloorWashington, DC 20005

Tel: 202-974-2400 Fax: 202-974-2410

h d

HUDSONI N S T I T U T E

$7.95

I SB N 978-1-55813-167-5

9 7 8 1 5 5 8 1 3 1 6 7 5

5 0 7 9 5

ISBN: 978-1-55813-167-5

comparing union sponsored and private pension plans

Documents