Will the Cadillac Tax Generate Revenue?

When the ACA passed in 2010, the Cadillac tax, a nondeductible 40 percent excise tax imposed on the portion of health coverage costs that exceeds $10,200 for single coverage and $27,500 for family coverage, was scheduled to take effect in 2018. It has since been delayed twice and is currently scheduled to take effect in 2022. At that point, the portion of health coverage costs that exceeds $11,200 for single coverage and $30,150 for family coverage will be subject to the tax.
The Cadillac tax is controversial. It is the first time that the historically unlimited tax exclusion for employment-based health benefits has been impacted. Although it has its enthusiasts, there has been bipartisan support for repealing it.
In May 2018, the Congressional Budget Office (CBO) estimated that the tax would generate $168 billion in tax revenue from 2022 to 2028. The Joint Tax Committee (JCT) and CBO assume that when employers reduce the comprehensiveness of health benefits to avoid the tax that they will in turn increase worker taxable wages such that total compensation is unchanged. Shifting the composition of compensation toward a higher proportion of taxable wages will translate into additional tax revenue. Repealing the tax would mean finding $168 billion (or the equivalent in today’s dollars) in new tax revenue.
Why is understanding the wage-benefit tradeoff so important? Research on the tradeoff has focused almost exclusively on what happens to worker wages when the cost of health insurance increases. The literature on the wage-benefit tradeoff is being used in such a way that it assumes that wage responses to health insurance cuts and health insurance cost increases are symmetric. Only one study on the wage-benefit tradeoff has been found that examines what happens to worker wages when the cost of health insurance decreases, as would be the case with the Cadillac tax. It examined the impact of community rating in New York State on the wages of older workers in small firms, firms that would see premium reductions as a result of community rating, and found that older workers in small firms saw their wages increase relative to workers in large firms and workers in other states. However, it is possible that wages do not respond symmetrically to increases and decreases in the cost of employment-based health insurance. There is evidence of asymmetric effects in other aspects of health care, labor markets, and elsewhere that could inform the economic theory on wage-benefit tradeoffs.
Even if workers’ taxable wages do not increase as a result of employers reducing the comprehensiveness of health benefits, the added tax revenue may still be realized. If employers kept the savings and those savings became corporate profits, they might become partially taxable as dividends and increased capital gains, thus producing higher federal tax revenue. If employers use the savings to purchase capital equipment, the seller of that equipment would see higher sales, which would result in federal tax revenue through a combination of profits, higher worker wages from sales bonuses, and worker wages possibly associated with the cost of installing the equipment. If employers used the savings to grow the business and hire additional workers, those workers would be paid wages that were taxable.

Figure 1 E Tw e B Rut Em c uc ndno ono o p le hm ? loy su d m g ue e re is v e L , t e lle t- ubo e s ha Ma y only s ha rs , t tTho ts v he k c e vov y e w long m a found e nd a S., rs C a g tOls a e a. nd k a 19 te on ha nd n C 9ta t9. " ( he $1 20 r aoline t 9,12 15 Fr le posit a inge )s V. C 7 for ftound ion th som Bre a fne e w a no a m fo e t fit m ily tre s a he d p v . loy ide c nd 20 o cost v e nc 17 e r tr s he e a . of we g t "ha e D D p , r e o a e tw m y H tc e ing e aon e ll b nd aaclt si h for e h I for d e lo e r s’ w he n r P ing su sa a tahe rlrt lta w a h b - n Ti r$1 a ie cg m e e s w 0,20 e ne e Re inc W fit efor r 0 a or rs c ee m a k and o e se ff s B m re s." s in r $2 c eoost ts fr e7,50 A dp e om b p D sponse y lie 0 t e cm a d ha h a w E rnd nge e c or sh onomi tk o for old e s in ins rre ’ s ow d s. Old c uc s E e t31 u ions n v rr eW a n if (nc 5 or i)n th e : k ers e Paul Fronstin is Director of the Health Research and Education Program at the Employee Benefit Research Institute Will the Cadillac Tax Generate Revenue? Employer Actions to Minimize the Impact of the Excise Tax c c wom om orkp per ere ns s he c b c 551 aont ost y ns tion SE i-s. v r56 ibut e , S? ne r3. d e"g ss eC a d oi:10.1 e r of d tnt he le e ess r m m for p of 0 alo x 80 im w Re ym /0 he um te 00 irtnt e he 36 m of -b re 84 a $3 tnt he sed 99 ,Re 45 p 32 he a sea 0 a y40 a m lt nd r02 e c h b n h a .t$6 e or tne , 90 igin Bos fit 0 f s. a ton tor e A d C t hose June w olle ith w g e w 20 or it W 15 h e k or e su k rm ing cr p on vloy eP tyra e ibu of p ee -only tr la ion N rg um s t e a e nd hr bm eoug rp fa loy 1m 7h p -ily e7. rs c a cht yorond v toll d p e:// ra uc g e ce d r te r uc .b in 201 d t cb ion .e y d tu/ he 8, orw the p- (E BRI). This Issue Brief was writte n with assistance from the Institute’s research and editorial staffs. Any views By Paul Fronstin, Ph.D., Employee Benefit Research Institute 1 A C em a m dp eill loy ria cc a e n He trca ont c x t on a hr etlnt rte ibu h P sh /utp old olic ions loa s w yd I m sns /2 ould ad t01 it eut 7/ not de ir 03 e found c b /tIe ly B m _1 to e t7 ha tt- . he 7.p t 16 he df 2 a p 0lt e 1h p r 5cent la 2n. 0of 16 E ec m ono Cp oloy nsm ide e ic r rin s r tghe fo ep ror 2 or 0 y 1t 7 e pd re tdha ictts t the ha yt w indiv ould idua inclrs r eae se cew iva ing ges t a high o offeset r e xpress See http eds :in t //whis ww r .c eb po or .gto a vr/e sy tshose tem/file of st? he fil ea = ut 2018 hor -a 06/ nd 53826 shoul-d he not althi bns e ura ascnc ribe ecd o vte o ra the ge. p offic df ers, trustees, or other sponsors 12 C Ale fe m we ns of , the Jeff a rb eov y, e a nd stuD da ievs a id lM. so loo Cutk le erd . 20 at 14 whe . "tW he ho r rip si ang ys f hor ea p ltublic h insu erm anc ploy e c eos e he ts le altdh co to a st c sha ?" nge Jour in hour nal of He s of altw h Ec orko or no tm he ic s t le he ve irl o re f e duc mtp ion in loyee he beanlt eh b fits a ene refit ps t aid ha at low occe urr rw ed ag te o taha void n th the ose C raedcill ea ivcing tax. a low Ae not r le he verl o suf e rve myp , loy cond ee uc be te ne d fit in 201 s, ass5 um as p ing a rt of EBRI, Employee Benefit Research Institute-Education and Research Fund (EBRI-ERF), or their staffs. Neither EBRI 2 See https://www.cbo.gov/sites/default/files/113th-congress-2013-2014/reports/45231-ACA_Estimates.pdf JanuaAry dd 3 or 1 ex, pa2 nd0 t1 oo9 ls t• o eN ncoo ur. ag e472 plan pa rticipants to be better consumers 73% 3% 14% 13 num She ine berr, of 38 L. w : 19 65 or 95 k -76 e . rH s e . e da m oi:10.1 ltp h C loyaer01 d e: C 6/ ost j.jhe s, W ale acgo.201 es and 4.04 Ag.0 ing: 08.A ssessing the Impact of Community Rating. Federal Reserve t of haat w hum ebina an c r, afound pital atnd hatot 5 p hee r rv ca ernt ia b of les t respond hat aff ee nt cts ha wor dk p er la w nne agd e s a to rinc e he reld ase const wora knt er w (Ro agsen es. 19 86). As a result of higher nor EBRI-ERF lobbies or takes positions on specific policy proposals. EBRI invites comment on this research. Introduction 3 Board. See https://www.cbo.gov/sites/default/files/113th-congress-2013-2014/reports/45231-ACA_Estimates.pdf health insurance costs, the theory is that employers will reduce worker wages in order to keep total compensation Add or expand incentives to engage employees in wellness programs Figure 2 53% 17% 14% Cred ? d y, John, Cutler aa nd nd N Ma icol da ris H an (1 era 99 ult 8) . 2 a00 nd 9. Gr Op ube tim r ( a19 l M 94 ar)g fou inal I nd nc tha om t ea n inc Tax Re rea fo se rm in he s: A a Mi lth insu crosim rul anc ate ion cost Ana s le lysi ad s. s t Re o sea an rch The excise tax on high-cost employment-based health plans (aka the Cadillac tax) is a provision in the 2010 Patient There Sugges arted e exa Cm itp ale tis w on:he Pa re ul Fr theonst wag in e. -b “W en ill e fit the tr a Cd ae doff d illac id Tanot x G e oc ne crur at. e For Rev in enue stanc ?” eE , B m RI a ny Iss e ue m p Bloy riee f, rs h no. alt 472 ed ( or E m su pspe loye nde e d unchanged (Summers 1989). However, employers may bear some of the costs of health benefits because they receive 4 Workers can pay out-of-pocket expenses on a pre-tax basis through FSAs and HSAs. Wi Summ ll e th rs, L P inc ap ra e ee w a r se Cad rN eum nc in hour e A b e H ve r. il 10 r 19 s w alac 80 ge 89 or , . A W k "e Som n Ta o d n r k ins u in e a g tlSim e x P P ar a d Ge e p p of le m er E ia Se u n inc cn m onomi rie se s, The ra er n a ca d se s of t E in t Univ e m Ma Re p he le nda o rnu ye sitty m v e r d of b Ce e B orMe en n ne of tu r lbo ifit w be or s." ur u? tk ne ie o A r, m n s e D s e e rm tip o cp a a H lo rn Ec tm SA yeed o nt s. nom , 2 of 0E i1 cc 8 ono Revie mw ic s. 79 A c (2) ce: ss17 ed7 - Protection and Affordable Care Act (ACA). When the ACA passed in 2010, the Cadillac tax, a nondeductible 40 percent 401(k) matching programs during the Great Recession. Workers did not experience a corresponding wage increase as Benefit Research Institute, January 31, 2019). a benefit from offering health benefits that they would not realize by paying workers strictly in cash wages; in those Implement or expand account-based CDHPs 57% 19% 7% 5 183. Octob er 26, 2015. http://fbe.unimelb.edu.au/__data/assets/pdf_file/0017/801170/1080.pdf excise tax imposed on the portion of health coverage costs that exceeds $10,200 for single coverage and $27,500 for a A rere su dluc t of tion the in cus ute in of csom ervp ic eenssa amo tion. ng One work e qrue s a st nion to e d their fami xamlie ine s ha is sw bhe rotahe derr w imp agle ic a gtions rowt h w for ill ov b era e ll highe pricers for . As the ovse eraw ll orkers By cases, Paw ul Fro orkers w nsould tin, not Ph.D. full, y Emp bear tloye he cos e Ben t of he eafi ltth b Re ene sefit as in e rch Iitns hetritut the e sh ort or long run (Garrett and Chernew ? Montgomery and Cosgrove (1993) and Buchmueller (1999) found mixed evidence as to whether employers Copyright Information: This report is copyrighted by the Employee Benefit Research Institute (EBRI). It may be family coverage, was scheduled to take effect in 2018. It has sinceE b m epl en d oye ee la -O ye nl d yt wice and is currently scheduled to d ov ema er tnim d e d ea cs c line om s, p pric are ed s fo to r w he or ak lth ers w carehose servic 40 es 1( okuts ) m idaet cohin f th ge perm og prloy am m w ent as una -baseff d esc ytsete dm . Istho isuld poss falible l as, w how ell, e tho veug r, t h ha the t Add or expand high performance networks, ACOs, or other delivery models 27% 3% 50% 2008) (O'Brien 2003). C Volpp utler, , K De e av xpa vin G id nd M., ., tG he ae nd or sha g Be rr igit L eow of tee tC ns he . tMa e w in, a or drkia for nd n. cD 19 e ae 98 vligib id . A "L le sc a b h. for or20 he M1a a 2. " rlt kh insu eC t hoosin Response ranc ge W . s t ise o lyRis : Liow ng -H Va ea lue lth Se Ins rv uic ra en s, U ce tCili os za ts: E tion, vide and nce used without permission but citation of the source is required. take effect in 2022. At that point, the portion of health coverage costs that exceeds $11,200 for single coverage and amount that prices may fall depends on the elasticity of supply of health care services. workers who lost 401(k) matches will never get offsetting wage or sala Crov y in erca re ge ases beF ca au m se ilyt he Cov ee cr onomi age c environment on Patie Hour nt C s Wo ost Sh rke ad ri." ng. RA " N JA DMA Jour 30na 8 (1 l o6) f E : c1 ono 635 m -16 ics3 29 6. d (3) oi:10. : 509 10 -530. 01/ja ma.2012.13616. 7 $30,150 for family coverage will be su Eb lim je inc at t et h o igt hhe cos tt a pla x. ns The tax was included in the ACA as a way to mitigate the rising a Muc t the h o tf th imee e e na arb ly le r de sea emp rc loy h e o rn th s to e h a wlta g or e -sbus ene pefit nd trta he de m off f atcound hing p 30% tha rog t rin am cr. ea 4% ses in health 37% insurance costs were mostly 6 Total Annual Premium A T A G L A N C E 6,272 1 8,054 ? Lubotsky and Olson (2015) found no evidence that school districts respond to higher health insurance costs by See http://www.cbpp.org/sites/default/files/atoms/files/cadillac_tax_letter.pdf Report Availability: This report is available on the internet at www.ebri.org cost of health care and to generate tax revenue to pay for other provisions in the ACA. borne by workers through lower wages or slower wage growth: Annual Employer Contribution to HSA 603 1,073 E berts, Rra enda ducll W., ing the and num Joe be A r . of Sttone eac. he 19 rs. 85 . "Wages, Fringe Benefits, and Working Conditions: An Analysis of 7 Increase employee cost sharing 42% 11% 16% The Morre e is de e ta viide lednc lite e ra tha tur t ea rst evrio eng ws e ca conomi n be focun env d in ir onm Qin e and nt c Ch an ha ernev w e (a 2014) n effe a cnd t on Lub wo he tstkhe y a r nedm Ol ploy son er (s w 2015) ill .c u t health ? When the ACA passed in 2010, the Cadillac tax, a nondeductible 40 percent excise tax imposed on the portion Total Annual Employer Spending 6,875 1 9,127 Compensating Differentials." Southern Economic Journal 52 (1): 274-280. doi:10.2307/1058920. The Cadillac tax is controversial. It is the first time that the historically unlimited tax exclusion for employment-based bene ?fit s. Eb In 2015 erts and , the Stone Pac ( ifi 19 c 85 Ma )r ifou timnd e Atss haotc a ian inc tion a rend ase Int in e ern m ap tion loye al Longsho r spending re on and heW altah insu rehourse anc Union e resu alg te rd ee in a d on a 8 Why is unde of he rsa tlt anding h cove trhe ag e tr a co dst eoff so s that im exc por eetd as $10 nt? W ,200 ith rfor esp si ec ngle t to c the ove C ra ag de ill a ac nd ta$2 x, th 7,5 e00 JC for T a nd famCily BO cov ass er uam ge e , tw ha ats when Gruber and Krue M ga ex r i( m 1991) um W aror gue ke tha r Ht SA the C on find trin ibut gs fro ion m their study w ould be ap 3 p,li4 c5 ab 0le to health ins 6 ura ,9nc 00 e if employers Table of Contents Reduce “richness” of plans 38% 31% health benefits has been impacted, albeit indirectly. Although it has its enthusiasts, there has been bipartisan support contract d tha ectr ew aould se in w cont ork inu ere w p arg ov es. iding For v e ervye rcy o m $1 p rin ehe crens ase ivein he healt ah insu lth insu rarnc anc e ec, ow ve or ra kg ee r w to ad goc es fe kwll b orky e r83 s. cIe ns ntts. ea d of scheduled to take effect in 2018. It has since been delayed twice and is currently scheduled to take effect in e Fr m ons ploy tin, P ers r ae ul. fo ducer tthhe com coing. mpr" eW he ha nstiv Deoe nes t sshe of Fu heta ur lth in e Hold sur a for nc e the in or Em dp elo r tyo ma ev nt oid -B a the sed t aH x, w ealtor h B ke erne wfit ag s Sy es w still e m inc ?"r e EaBsRI e, were mandated to provide coverage much like they are mandated to provide workers’ compensation coverage. Total Subject to Cadillac Tax $ 9,722 $ 24,954 Introduction .......................................................................................................................................................... 3 for repealing it. Repealing the tax in a budget-neutral fashion would mean finding the estimated $168 billion (or the reducing the comprehensiveness of health coverage, as the Cadillac tax was expected to do, the five-year contract was with about I 20 ss 22 75 ue. p B A e rie r t cf te ha ( nt Et m of pp oint loy thee , e $ the 168 Be p ne or bfit ill tion o ion ci Resea f he terdc a h I altbh co ov nst ei v tc ut eore m a)ging .e cos from ts tinc hatr e ea xc sed ee d inc s $11 ome ,200 and fo pra y sirn oll t glea cx r ove ev re anue ge a nd associated 9 Reduce spousal subsidy or implement spousal surcharge 37% 15% 13% 14 ? Gruber and Krueger (1991) studied the impact of the workers’ compensation program on worker wages and Community rating requires that insurers offer health insurance policies at the same price within a geographic region to all equivalent in today’s dollars) in new tax revenue to replace the amount of revenue that the Congressional Budget Office Why Use a Cadillac Tax .......................................................................................................................................... 3 expected to trigger $60 million to be paid toward the Cadillac tax. Other recent examples of compensation and work $30,150 for family coverage will be subject to the tax. with the increase in worker wages. One of the potential shortcomings of the assumption relates to the fact that 8 1 persons re found gardle t Sha sou s to rf w cthe e or : k E ire B rhe R w Ia e a lth s gte is m s w ta atus te ers e . us 86 i ng pea rc ve ent ra low ge pr ere a m s ia um re s ault nd . a v erage employer contributions to (CBO) estimated the tax would generate from 2022 to 2028. Increased tax revenue is expected to come from two e Gnv arr ireon tt, m Be ow nt een, nha and nceMic meha nte s, p l Cr he esu rne mw ab . l20 y b 08 ec. a"u H se eaof lth I the nsu strraong nce e acnd onom Labic o rc lMa imr ak te et , s: inc Clu onc dee unl pts, Op imiteedn Q paue id st faions mily, le and ave research on the wage-be M ne ove fit to t fu ra ll d ree poff ha lacemens fo t of C cDH usPe sd tr aa te lm gyost exclusively on what happens to worker wages when the cost A Look at the Wage-Benefit Tradeoff ................................................................ 33% ................................ 25% ...................... 4 HSAs from Exhibit 8.7 in https://www.kff.org/report-section/2018-employer-health- sour 10 ces. One source is the direct effect of the tax. Health plans that exceed the tax thresholds will incur the ? D The ataC N ae de illdas." c ta Inquir x is cy o nt 45 (1 rove)r:si 30 al. -57. It is the first time that the historically unlimited tax exclusion for for certain workers at high-tech companies such as Adobe, Facebook, Google, Microsoft, and Netflix, and, among other See http://www.cbpp.org/sites/default/files/atoms/files/cadillac_tax_letter.pdf of health insurance increases. The literature on the wage-benefit tradeoff is being used in such a way that it assumes 1% ? Gruber (1994) and Sheiner (1995) found that as a result of an increase in health insurance costs, worker The Relationship Between Benefit Costs and Other Business Expenses: Will the Tax Revenue Materialize? ................... 7 15 benefits-survey-section-8-high-deductible-health-plans-with-savings-option/. nondeduc et m ible ploy 40 m e pnt er-cb ea nt sed ta x he on altth b he evne alue fits ha of ts b he e ce ov n im erap ga ec a te bd ov . eA tlthe houg thrh it esh old has it . The s ent ot hus her ia sou sts,r c the e is re t he ha s b indir eee n ct effect. things, relaxation of dress cM od ove es a to at dW efin al em d ca or nt t. ribu tion model 10% 22% that wage responses to health insurance cuts and health insurance cost increases are symmetric. Only one study on 11 wages grew slower over time among particularly more costly groups, such as older workers, workers with As cited in Volpp, Lowenstein and Asch (2012), a series of studies found that increasing and decreasing patient copayments Gruber, Jonathan. 1994. "The Incidence of Mandated Maternity Benefits." American Economic Review 84 (3): 622-641. The Joint Tax Committee (JCT) and CBO assume that when employers reduce the comprehensiveness of health Conclusion bipa ................................ rtisan support for rep ................................ ealing it. ............................................................................................ 8 the wage-benefit tradeoff has been found that examines what happens to worker wages when the cost of health family coverage, and women of child-bearing age. for prescription medications do not have mirror-image effects. Raising copayments reduced use of prescription drugs and 0% 20% 40% 60% 80% 100% benefits to avoid the tax that they will in turn increase worker taxable wages such that total compensation is insurance decreases, as would be the case with the Cadillac tax. This paper examined the impact of community rating References ............................................................................................................................................................ 8 Gruber, Jonathan, and Alan B. Krueger. 1991. The Incidence of Mandated Employer-Provided Insurance: Lessons from medication adherence, but reducing copayments did not increase use of prescription drugs and medication adherence nearly 2 ? In May 2018, the Congressional Budget Office (CBO) estimated that the tax would generate $168 billion in tax The Relationship Between Benefit Costs and Other Business Expenses: Will the Tax unchanged. Shifting the composition of compensation toward a higher proportion of taxable wages will translate into in Ne ?w YP or ak uly st a atnd e on Hetrhe ring w( a1 g99 es of 9) fou oldnd er w tha ortk p er re s din sm icted am ll e fir dm ica sl e , fix rm pes t nsh ea s ha t would d a ne see ga tp iv re em im ium pa crte on ductthe ion w s a ag s ea s of resold ult e of r Conclusion as much. W Oth orke er rs' exa Cmp omle pe s ns ara et io asn I fons llow ursa : nCr ce e.d Vol. 5, dy and in HeTa rault x P (olic 2009) y a nd foun the d tha Ect onom margyina , bly w D ea lfa vid re B cha rad ng for esd , fo11 r th 1- e14 Aus 4. The tralia n Endnotes ............................................................................................................................................................. 11 revenue from 2022 to 2028. The Joint Tax Committee (JCT) and CBO assume that when employers reduce the Reve Snue ource: NMater ational Buial sineize ss Gr? ou p on Health (2015). additional income9 and payroll tax revenue. In a 2014 report, the CBO found that roughly one-quarter of the increase in communiw ty or rk ae ting rs. (Adams 2007). Adams found that older workers in small firms saw their wages increase relative to The inco me Ca d ta ill MI x as c T P truc tarxe ture ss is . c ur A ac re rce e no nt ssed ly t ssc y mme Oc hetd ob ule tric erd w 23 tith o , t20 a re ks 15 ep e e . c ff ht t eto c tp t :// inc in 2 w re02 w aw s2. e.nb s a Bnd eerc .or a dus egc/c e re h tahe a sp e s tte h in rrs/ eta sh c11 x old ra 27 te s0 st .ha Brow t trigg n e ar n d the Yüc ta ex a l (2000) re tie d to comprehensiveness of health benefits to avoid the tax that they will in turn increase worker taxable wages tax revenue stems from excise tax receipts, and roughly three-quarters is from the effects on revenues of changes in E ven if workers’ taxable wages do not increase as a result of employers reducing the comprehensiveness of health workers in large firms and workers in other states as a result of community rating in New York. A similar study reference much of the literature that finds asymmetry between gasoline and crude oil prices such that gasoline prices rise inflation, if the tax takes effect in 2022, the 40 percent excise tax will be imposed on the portion of health coverage 3 such that total compensation is unchanged. Shifting the composition of compensation toward a higher e bm ene ploy fits, t ees’ het aaxa dd be le d cto ax r mpeevns enue ation ma and, y st ill to bae le re ss ae liz r eed x. te Inf e t, m in e ploy mp eloy rs reerd s’ uc de e dtuc het ible cost e of xp he ens ae lts. h b ene fits and they do not ? Olson (2002) found that health insurance reduced wages by only 20 percent among married working women. K exa olst maine d, d Jona how tha the n T., mor and e ge Ane ma ra nd l m a o Ev . e Km ow ent al sk tow i. 20 ard 12 lim . "iMa ted nda pre te m -B ium ased va H ria etaion in lth Retfor hem sm and all g the roup La b mor ar k Meatr a ke dtop : E te vd ide bnc y e more quickly when the price for crude oil increases, as compared to gasoline prices declining when crude oil prices fall. costs that exceeds $11,200 for single coverage and $30,150 for family coverage for that year. While the tax is proportion of taxable wages will translate into additional tax revenue. Repealing the tax would mean finding increase worker wages, where does the money go? If employers keep the savings and those savings become from the Massachusetts Reform." NBER Working Paper No. 17933. many states in the early 1990s affected the wage gap between workers in small and large firms. It also examined 12 controversial for a number of reasons, a main point of contention is whether the tax will generate federal tax revenue Figur ? e$ Mille s 168 r , bJr ill. ion (or (2004) t he found equiv thaale t n ha t in t ving od he aya’s d lth insu ollars) ra nc in ne e lew d tta o x r a e 10 ve –nue 11 p . e rcent wage reduction among prime-aged See http://www.americanhealthpolicy.org/Content/documents/resources/AHPI_Excise_Tax_October_2015.pdf corporate profits, they might become partially taxable as dividends and increased capital gains, thus producing higher whether there was an increase in employment of older workers, as the marginal cost of hiring an older worker fell as While economic theory appears straightforward, how the Cadillac tax plays out in the real world is a subject of intense as predicted. The tax is expected to generate tax revenue because as employers cut back on the generosity of health Why Use a Cadillac Tax working men. Figure 1, Employer Actions to Minimize the Impact of the Excise Tax ........................................................................ 4 13 L fe ubo derta sl ta ky, x Dra ervre enue n, a. nd If e Crm aig ploy A. eOls rs us on. e 20 the 15 s. a" vP ings remiu to m p ur Co cp ha ay se me ca np ts a ital e nd qtuipment he Trad, e- the Off selle Betw r e of en Wa that g eq es a uipment nd Em w ploy ould er- See http://ushealthnews.mercer.com/article/462/introducing-mercer-s-excise-tax-survival-kit premiums were less dependent on the age composition of a firm’s employee pool (Rutledge and Crawford 2017). The deba ?t e. W Que hy st isions unde arris se tanding relate d the to :w age-benefit tradeoff so important? Research on the tradeoff has focused almost benefits, workers will receive higher wages in return. While it is debatable as to whether higher wages will materialize, The theory behind capping the health exclusion rests on the assumption that, because of the ability of workers to see highe Prr ov saide les, w d Hhic ealt h w h Ioul nsu dr a re nsu celt ." in fe Jourd na erl o al t f a Hx r eae ltvh Ec enuono e thr m oug ics h a 44: c63 om -79 bina . dtoi:10.1 ion of p 01 rofit 6/js, hig .jheale he co.201 r work 5.08 er w .0 ag 06 es f . rom study found that the earnings gap fell equally for workers of all ages, but the premium restrictions had only a small 14 The highe m r or tae e x xc rre e lus cveent inue ve fin ly m d on ing ayw s a st ha ill ls t a o hp ag p pe e p ne a erns rif t a lly to he su w sor p ap v kor ings f ert w a aw rgom a eg s w e le -b he ss ene n th ge fit ne e t rrcous he aost de off, a of a he ltnd h b alth insu tehe neyfit gr s a o anfur rcee t us he inc err d e t o a for ses. e xa otm he The ine r p u a lit rr e pa rose nge atus t r e of ha on t Fi g Sur eee http 2, A :/ v/ew ra wg w e. w As nnu j.coa m/ l Parrtic emle ium s/w s a esnd t-coE am stp -p lo oy rte-rc o Cntra ontr cib t-ma utions y-stick to -H em SA ps, 201 loyers-8 w ................................ ith-cadillac-tax-1432745959 ......................... 8 exclude the value of employment-based health coverage from taxable income without limit, workers prefer additional sales bonuses, and worker wages possibly associated with the cost of installing the equipment. If employers use the ? Whether reducing the comprehensiveness of coverage will reduce use of services among the small portion of effect on employment of older workers. the wage-benefit tradeoff is being used in such a way that it assumes that wage responses to health insurance r m eor sue lt spe in tc aifi xa cb e le ff e inc cts om . H e.ow ever, the findings are more mixed than earlier work: Mille 15 r, Jr., Richard D. 2004. "Estimating the Compensating Differential for Employer-Provided Health Insurance." health insurance benefits over cash wages (and other forms of compensation) — and because of this preference, they savings to grow the business and hire additional workers, those workers would be paid wages that are taxable. See http://www.businessinsider.com/gates-foundation-announces-unlimited-leave-policies-2015-10 and the population that accounts for most of the spending. cuts and health insurance cost increases are symmetric. Only one study on the wage-benefit tradeoff has been International Journal of Health Care Finance and Economics 4 (1): 27-41. are “over-insured.” Once workers either have health coverage or see an increase in the comprehensiveness of their D http espit ://mo e the nerye .c b nn eing .com only /201 one 5/06/ st04/ udy o new n w s/cha omp t h aa nie pp se /w ns a lma to w rtor -jekaenrs w -mus ageic s w -tehe mp n th erature e cos /in t dof ex.h hetml alth ins ?iid=ur SF_L anc Ne decreases, ? Baicker and Chandra (2006) found that rising health insurance costs were fully offset by reductions in workers’ found that examines what happens to worker wages when the cost of health insurance decreases, as would be While most versions of the benefit-business expense tradeoff result in higher federal tax revenue, there are exceptions. cove ? r ageT , he the im the plic or ay t ions is the of y rtehe duc n use ed us m eor of e he hea alt lth c h ca arre e ser serv vic ice es s t . han they otherwise would. Workers over-insure economists continue to predict that when employers reduce the comprehensiveness of health insurance, worker wages 16 Refere wnce ages. s See http://ushealthnews.mercer.com/article/462/introducing-mercer-s-excise-tax-survival-kit Montgom the ery , ca M se arw k, ita h nd the Ja C m ae ds C illaosgr c taxov . e I. t 19 exa 9m 3. " ine The d the Eff im ectp a of ctE of mp cloy omem e uni Betne y rfit ats on ing in N the e D w e m Yor and k St for at e P aon rt-Ti the m e w ages of Some employ 10 ers report that they were inclined to use the savings from offering reduced health coverage to add health- because health insurance premiums are not included in taxable income, while out-of-pocket spending on health care will increase. It is possible that the effects are asymmetric. In other words, it is possible that wages do not respond 16 Adams, Sc old Wot or etk r. ew 20 rs." or 0k 7. " I endus rs in s Hea trltim a h l a aIll fi ns nd u rm rLaa s, fir nbcor e Ma m Re s t rla kh teion a t tRe w s Re ould forv m ie see a wn d 47 p E r m e (1) m p: ium loy 87 e -e r98. e C dom uc tp ioens ns a ats a ion: reThe sult C of ase com of m Puni uret yC r om ating, muni atnd y 17 4 ? The extent to which the tax will result in increased federal tax revenue. related benefits not subject to the tax, such as dental or vision benefits, or to add to the 401(k) match. If employers Shifting compensation from nontaxable health benefits to taxable retirement benefits may result in increased federal income services does not automatically enjoy the same tax preference. As a result, workers are theorized to prefer higher- ? Bhattacharya and Bundorf (2009) found a significant wage differential between obese and non-obese women symmetrically to increases and decreases in the cost of employment-based health insurance. There is evidence of Rating in New York." Journal of Public Economics 91 (5-6): 1119-1133. doi:10.1016/j.jpubeco.2006.09.010. found that older workers in small firms saw their wages increase relative to workers in large firms and workers increased contributions to worker retirement savings plans, higher tax revenue would not be realized until the worker tax revenue, but the increased revenue may occur outside of the budget window for which legislation is scored. premium comprehensive coverage with low cost sharing over lower premium, less comprehensive coverage with higher asymmeta rm ic ong effe c tho ts in o se w the ith insu r aspe racn ts of ce, bhe uta no lth c diff are er, ela nt bia ol fo r mra r tk hose ets aw nd ithout else w ins he ur re a nc tha e. t c ould inform the economic theory O'Brien, Ellen. 2003. "Employers’ Benefits from Workers’ Health Insurance." Milbank Quarterly 81 (1): 5-43. This paper addresses this last consideration. The 17 next section examines the literature on the wage-benefit tradeoff. in other states. However, it is possible that wages do not respond symmetrically to increases and decreases in 11 took a distribution at some point in the future. Similarly, if employers adopt an HSA-eligible health plan and recycle c 18ost sharing. Ultimately, higher-premium, lower cost-sharing health coverage leads to overuse of health care services, on wage-benefit tradeoffs. It is also possible that the effects vary by industry and geographically. Ana Wnd, ork eP r rc iy oa ntrib nka. ution 2013 s . to" The HSAs E ff outs ectid of e o Ris f ping ayro Hlle d ae ltd h I ucns tion ura dnc o no e C t o cst oun s on t toC wom ardp the ens a Ctaion dill aacnd tax Em thr pe loy sho mld e.nt ." Accessed The following section discusses why the wage-benefit tradeoff may be irrelevant when it comes to predicting the impact the cost of employment-based health insurance. There is evidence of asymmetric effects in other aspects of some ? or K aolst ll of atd he a nd savK ings owafr lso ki (20 m the 12 highe ) found r de tha duc t tw ible ag eint s a o dw jus ortk ee dr s ful ’ H lyS tAo s ( offs ass eu t m the ing cos the t of com the bina heta ion o lth ins f th ure a nc pre e m miu am nda te which drives up insurance premiums and makes coverage less affordable. Olson, Craig A. 2002. "Do Workers Accept Lower Wages in Exchange for Health Benefits?" Journal of Labor Economics October 26, 2015. http://aida.wss.yale.edu/~pa88/research/docs/PAnand-JMP.pdf of the Cadillac tax on federal revenue. health care, labor markets, and elsewhere that could inform the economic theory on wage-benefit tradeoffs. A and s mte he nt io e tha m ne p td loy t ook e ae br ov ce on ff, eQin a tcrtibu wit tn io h th dn to th Ce he 20 rne e 0 6 Ma w HSA (20 ss d14 o ac not )h found us te rt igg ts la te ha rw t the . w or Ck ae drill w ac a g ta ex) s fe , tll 15 p he HSe Ar c ce on ntt rw ibut heion w n the ill c ost not of gehe nearla th te tax 20 (2, Part 2): S91-S114. Employers have been expected to take a number of different actions to avoid the Cadillac tax (Figure 1), including revenue unless a worker takes a distribution not used for a qualified medical expense. Finally, more employers may insurance increased, but also found that the tradeoff was larger during the Great Recession of 2007–2009. They Baicker, Katherine, and Amitabh Chandra. 2006. "The Labor Market Effects of Rising Health Insurance Premiums." ? Even if workers’ taxable wages do not increase as a result of employers reducing the comprehensiveness of ? Clemens and Cutler (2014) found that only a small fraction (15 percent) of increased health and retirement reducing the comprehensiveness of employment-based health coverage. This would cause cost sharing for health care Pauly, Mark V., and Bradley Herring. 1999. Pooling Health Insurance Risks. Washington, DC: AEI Press. c eonc xpalude nd p a th id atle e acvono e pr m og icr a em nvs. ir onments matter when it comes to enabling or precluding the wage-benefit tradeoff. Journal of Labor Economics 24 (3): 609-634. doi:10.1086/505049. A Look he at alt h b the ene Wa fits, t ge he- aB dd eene d ta fi x r t e Tr veade nue m oaff y still be realized. If employers kept the savings and those savings benefit costs were offset through a reduction in wages among public school workers, though the effect was not services (such as deductibles and copayments) to increase, translating into a reduction of use of health care services Hence, employers may be able to adjust worker wages down (or slow wage growth) in economic climates that include became corporate profits, they might become partially taxable as dividends and increased capital gains, thus Economists assume that when employers reduce the comprehensiveness of health benefits they will in turn increase Qin, P In dete ar ige m stini , at ang is nd ticw Mic ahe llyha td he iff erl C e a re n H he ntr ne SA thw a -e n ze . ligib 20r 14 o. le. he "Ca om lth p pela ns n tr ating iggW era s t ge he D C iff ae drill ea nt cia ta lsx, e and mtphe loy Iem r p aa nd ct e of mH ploy eale th I e cn ont surra ib nut ceions in t he via among workers and their families. Health care costs and premiums for employment-based health benefits would Bhattacharya, Jay, and M. Kate Bundorf. 2009. "The Incidence of the Healthcare Costs of Obesity." Journal of Health recessions and high unemployment rates, but during a strong economy and low unemployment rates it may be more 18 producing higher federal tax revenue. If employers use the savings to purchase capital equipment, the seller 5 worker wages such that total compensation is unchanged. Hence, if employers reduce health benefits to avoid a tax Public Sector on Wages and Hours." Journal of Health Economics 38: 77-87. payroll deduction to the HSA are added to premiums. Worker contributions would be included because under the tax decline aE s a cono resu mic lt sof 28 the (3) r: ed 64 uc 9t-ion in 658. health care services, while workers’ taxable wages would increase. Some difficult for an employer to shift the rising cost of health benefits onto workers. Similarly, when the cost of ? Qin a of thand t eC quipment hernew (w 20 ould 14) see also highe found r sa a 15 les, w perhic cent h w re ould duc trie on sua ltm in fe ongd p eublic ral ta sec x re tor ve nue wor k th er roug s, tha h a t tc he om rb ed ina ucttion o ion f cap, workers’ taxable wages are assumed to increase. Shifting the composition of compensation toward taxable wages code wordkoi:10.1 er cont01 ribu 6/tjions .jhea a le re co.2014 treated .08 as e .001 m.p loyer contributions. As a result, worker contributions could trigger the employers, however, will choose to pay the Cadillac tax. Employers offer health coverage to their workers to be employment-based health insurance falls, employers may have to increase wages when the economy is strong and w pra ofit s la s, hig rger he dur r ing wortkhe er g wra eg ae t sr efr com ess ion o sales b f 200 onus 7–e2009 s, and , aw nd ortk ha ert w it aw ga es l s p aoss rgeibl r a ym aong ssoc non iate- duni wit on h th wor e k cost ers. of will translate into additional tax revenue. This is a long-standing position of economists and others, and is one of the Brown, Stephen P.A., and Mine K. Yücel. 2000. "Gasoline and Crude Oil Prices:Why the Asymmetry?" Economic and Cadillac tax. However, it is unlikely that HSA-eligible health plans would have triggered the Cadillac tax in 2018. competitive in the labor market (Fronstin forthcoming). To the degree employers are concerned about the impact that unemployment is low in order to compete and retain workers, but when the economy is weak and unemployment is Rosen, Sherwin. 1986. "The Theory of Equalizing Differences." Handbook of Labor Economics 641-692. installing the equipment. If employers used the savings to grow the business and hire additional workers, reasons why a group of economists signed a letter to congressional leaders in opposition of any action that would Financial Review: Federal Reserve Bank of Dallas. Accessed October 26, 2015. Premiums for HSA-eligible health plans averaged $6,272 for employee-only coverage and $18,054 for family coverage cutting health benefits may have on recruiting and retaining workers (even with a compensating wage income), they high, employers may not need to increase worker wages. The bottom line is that employers pay higher compensation EBRI Issue Brief is registered in the U.S. Patent and Trademark Office. ISSN: 0887 ?137X/90 0887 ?137X/90 $ .50+.50 ? Anand (2013) found evidence that rising health insurance costs were shifted onto workers through hig 6 her those workers would be paid wages that were taxable. weaken, delay, or reduce the Cadillac tax, unless an effective alternative to the Cadillac tax was first enacted. https://www.dallasfed.org/assets/documents/research/efr/2000/efr0003b.pdf in 2018 (Figure 2). Employer contributions averaged $603 and $1,073 for employee-only and family coverage, may choose to continue to pay the Cadillac tax. when comp © et2 it0 ion fo 19, Erm sc plo ary ce ee la B benef or isit high, Resea arc nd h In thastt c itom utep ?eE ns du atca ion o tion fta end n inc Resea ludes t rch F ax-un fred. e he All r alth b ighe tsne re fit ser s. ved. employee-paid premiums, but not through lower wages or by affecting other employee benefits. respectively. In the absence of worker HSA contributions, premiums and employer contributions totaled $6,875 for e e e e e e e e e eb b b b b b b b b br r r r r r r r r ri. i. i. i. i. i. i. i. i. i.o o o o o o o o o or r r r r r r r r rg g g g g g g g g g IIIIIIIIIIs s s s s s s s s ss s s s s s s s s su u u u u u u u u ue e e e e e e e e e B B B B B B B B B Br r r r r r r r r rief ief ief ief ief ief ief ief ief ief • • • • • • • • • • J J J J J J J J J Ja a a a a a a a a an n n n n n n n n nu u u u u u u u u ua a a a a a a a a ar r r r r r r r r ry y y y y y y y y y 3 3 3 3 3 3 3 3 3 31 1 1 1 1 1 1 1 1 1,,,,,,,,,, 2 2 2 2 2 2 2 2 2 20 0 0 0 0 0 0 0 0 01 1 1 1 1 1 1 1 1 19 9 9 9 9 9 9 9 9 9 • • • • • • • • • • N N N N N N N N N No o o o o o o o o o.......... 4 4 4 4 4 4 4 4 4 47 7 7 7 7 7 7 7 7 72 2 2 2 2 2 2 2 2 2 10 11 2 6 8 7 5 9 4 3 A research report from the EBRI Education and Research Fund © 2019 Employee Benefit Research Institute