On the Replacement Adequacy of Canada s Retirement Income System is there a problem? what if the C/QPP were expanded would expansion of private savings help inter-generational fairness more on the age of retirement Michael Wolfson, uottawa 1
Acknowledgments and Caveats huge thanks to the LifePaths team in Statistics Canada s Modeling Division financial support over many years from Policy Research Initiative and Federal HRSDC full responsibility for the range of assumptions used in this analysis, and interpretation of results Wolfson LifePaths is public domain; to do further simulations, feel free to contact Statistics Canada ( caveat: LifePaths is a complex simulation model, some parts of which are relatively new ) 2
Criteria for an Adequate Retirement Income System avoid poverty / low income support continuity of consumption / maintenance of living standards into retirement provide safe pensions = reduce uncertainty (or at least share risks fairly amongst individuals, employers, fellow employees, taxpayers) 3
Basic Conclusion re Is there a problem? about half of the baby boom population (those born between 1945 and 1970) in the middle 50% of the earnings distribution for their prime working age years (i.e. ages 40 to 65) can expect a decline in their net replacement rate (RR) / consumption possibilities after retirement of at least one-quarter (see: http://www.irpp.org/pubs/irppstudy/irpp_study_no17.pdf) 4
What is the Net Replacement Rate (RR)? ratio of post-retirement consumption possibilities to those prior to retirement norm or objective: 100%; i.e. ability and likelihood of maintaining pre-retirement living standards after retirement consumption possibilities gross income less income and payroll taxes less savings plus dis-saving (i.e. running down assets) 5
Comments on Net Consumption Replacement Rate (RR) not the same as gross income RR or net income RR income RRs are easier to estimate and analyze, but not as close to the desired concept of consumption possibilities next slide shows basic accounting and then a slide showing the joint distributions 6
Replacement Rate (RR) Adequacy Basic Accounting Pre-Retirement Post Retirement add earnings public pensions imputed rent on owned home subtract income and payroll taxes income taxes RPP, RRSP and RRIF income imputed rent on owned home net withdrawals of home equity RPP and RRSP contributions mortgage principal payments divide EAU (= equivalence scale) EAU (= equivalence scale) result consumption pre-retirement consumption post-retirement not included: work-related expenses, other investments (income, saving, dissaving), consumer durables, business assets, inheritances and gifts inter-vivos 7
Analytical Method LifePaths microsimulation model of the Canadian population and its interactions with the tax / transfer system longitudinal dynamics explicitly incorporated, based on integration of analyses of many data sets including censi, LFS, SFS, FHS, T1, PPIC, etc. validation against historical data, external peer reviews for selected modules as if ideal longitudinal microdata for huge sample (millions) of the Canadian population both for the status quo and for selected what if scenarios results then simply cross-tabulations of synthetic but realistic full lifetime longitudinal population samples 8
Gross versus Net Replacement Rates females > 150 130-140 115-120 105-110 males > 150 130-140 115-120 105-110 95-100 85-90 75-80 65-70 55-60 45-50 35-40 25-30 15-20 5-10 95-100 85-90 75-80 65-70 55-60 45-50 35-40 25-30 15-20 5-10 < 1 < 1 5-10 15-20 25-30 35-40 45-50 55-60 65-70 75-80 85-90 95-100 105-110 115-120 130-140 > 150 < 1 9 for realistically heterogeneous populations, a 60 to 70% gross RR translates into a 60 to 100% net RR, so that gross RRs are a poor proxy; better to do analysis directly in terms of net RRs
Prime Age (40 65 ) Earnings percent ~ 50% indiv s 22 females 20 18 19-27% males 16 14 19-26% 12 10 8 6 4 2 0.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 125-150 >150 earnings age 40 65 $000s 10
Average Net Replacement Rates by Prime Age Earnings Main Result (1960-65 cohort) % 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0 females males replacement rate declines with pre-retirement earnings for the middle 50% of earners in this birth cohort, average decline in consumption possibilities is in the 15 to 35 percentage point range, with declines ~5 7 percentage points higher for men ~ 50% ~ 25%.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 earnings age 40 65 $000s 125-150 >150 11
Average Net Replacement Rates by Prime Age Earnings by Home Ownership % 180 160 140 120 100 80 60 40 housing ignored imputed rent only 50% imputed rent + 50% annuitized no imputed rent + 100% annuitized 20 0 ~ 50%.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 125-150 >150 earnings age 40 65 $000s 12
Average Net Replacement Rates no housing, males, age 70, by earnings base and discounting 180 160 140 120 % Best 5-70 AW Prime - 70 AW Updated Career - 70 AW Best 5-70 CPI Prime - 70 CPI Updated Career - 70 CPI 100 80 60 40 20 0 ~ 50%.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 125-150 >150 earnings age 40 65 $000s 13
% 180 160 140 120 100 80 60 40 Average Net Replacement Rates no housing, females, age 80, by earnings base and discounting Best 5-80 AW Prime - 80 AW Updated Career - 80 AW Best 5-80 CPI Prime - 80 CPI Updated Career - 80 CPI 20 0 ~ 50%.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 125-150 >150 earnings age 40 65 $000s 14
% 100 90 80 70 60 Fractions (%) with Net Replacement Rates Below 75% by Birth Cohort both sexes, AW discounting, 50-50 housing fraction with net RRs falling by at least one-quarter is increasing for successive birth cohorts about half those with middle range incomes can expect a decline of at least one-quarter 50 40 30 20 10 ~ 50% 1945-50 1950-55 1955-60 1960-65 1965-70 0.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 125-150 >150 earnings age 40 65 $000s 15
Fractions (%) with Net Replacement Rates Below 75% or Above 100% by Discount Rate % 100 both sexes, 50-50 housing <75% AW 90 80 70 <75% CPI >100% AW >100% CPI 60 50 40 30 20 10 0.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 125-150 >150 earnings age 40 65 $000s 16
Houston, do we have a problem? net replacement rate is correct concept to use for assessing adequacy in terms of maintaining living standards / continuity of consumption this concept is difficult to operationalize results are sensitive to: birth cohort, earnings base, discount factor, treatment of housing, age 70 versus 80 nevertheless about half of middle (50%) income baby boomers will face a decline in net consumption of at least one-quarter 17
Private Sector Options DB RPPs checkered history aggressive actuarial assumptions opposition to early 1980s excess interest indexing tension between limiting tax deferral (Income Tax Act) and assuring adequate financial security for plan members (Pension Benefits Standards Acts) Tax Assistance incentives poorly targeted compared to adequacy gaps recall 1983 Parliamentary Ctte target middle not upper incomes DC Plans (both RRSPs and RPPs) Ambatscheer ideas much lower MERs, phased entry to annuity markets (if not RRIFs) 18
Public Sector Options / Challenges obvious options modest expansion of CPP (and QPP) but what about long run indexing of OAS / GIS / SPA / income tax raising age of entitlement to public pensions / normal age of retirement inter-generational fairness health care represents an even larger unfunded liability than public pensions 19
Expansion of Mandatory DB Benefits double C/QPP increase nominal replacement rate by 25%, no change to YMPE (or to YBE) wedge increase replacement rate to 40% starting at ½ AW and extend up to 2 AW cost proportional to 5.2% payroll for future service cost of current C/QPP retirement + post-retirement survivor post-retirement benefits double double current C/QPP wedge question: which option better targets any problem? 20 UCAE = updated career-average earnings per C/QPP
Average Net Replacement Rates both sexes, AW discounting, 50-50 housing, 1960-65 cohort % 200 180 160 140 120 Base Case Scenario CPP Doubled YMPE Raised and 40% Wedge OAS/GIS etc AW Indexed 100 80 60 40 20 0 ~ 50%.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 125-150 >150 earnings age 40 65 $000s 21
% 100 90 80 70 60 50 40 30 20 10 0 Fraction (%) with Net Replacement Rate below 75% by Reform Scenario both sexes, AW discounting, 50-50 housing, 1960-65 cohort Base Case Scenario CPP Doubled YMPE Raised and 40% Wedge OAS/GIS etc AW Indexed ~ 50%.1-15 15-25 25-35 35-50 50-65 65-80 80-100 100-125 125-150 >150 earnings age 40 65 $000s note variation in targeting of increases by earnings 22
What About Tax Incentives for Private Savings? For the same dollar of contribution to a registered pension plan or RRSP, the current tax system provides a larger benefit to individuals in higher tax brackets. Some witnesses found this inequitable. More important, the fact that the tax benefit is worth more to those with higher incomes means that its effectiveness in encouraging people to save for retirement is strongest in the upper income groups and much weaker in lower and middle income groups. We feel that there should be a reasonable limit on the maximum private pension that can be accrued with preferential tax treatment (effectively up to full replacement at about the income of a high school principal) Report of the Parliamentary (Frith) Task Force on Pension Reform, 1983, p64 23
What About Increasing Private Savings? (from Kevin Moore, CEA, June 2011) (TER = total expense ratio) 24
What About Inter-Generational Fairness? (I) We have concluded that something along the lines of wage indexing is necessary for public pensions. However, we are not sure that the AIW is the best indexing factor. An appropriate adjustment factor should be simple and easy to understand. It should also have characteristics such that benefits would be higher when real average wage growth is higher, labour force participation is higher, unemployment is lower, or the old-age dependency ratio is lower, and would be lower when the opposite circumstances prevailed. Report of the Parliamentary (Frith) Task Force on Pension Reform, 1983, p17 25
What About Inter-Generational Fairness? (II) To portray America as riven by generational warfare, young against old, is therefore an exaggeration. Worse, it obscures a deeper divide, of class rather than age. The big problem of the American welfare state is not that the old get too much, but that the rich do. (Economist, January 11, 1997) 26
While there is great concern about the effects of increasing longevity on pension costs and hence on inter-generational fairness, our simulations suggest that for quite a wide range of life expectancy scenarios, this has a much smaller impact than the strength of the economy. notwithstanding widely expressed concerns regarding inter-generational fairness, differences within generations are far larger than those between generations. Women s net lifetime transfers minus taxes are hundreds of thousands of dollars greater than those for men, while the differences between the poor and the rich within any given generation are larger still. 27 (Wolfson and Rowe, 2007, JEI)
LifePaths Simulation of Sources of Variance in Net Present Values of Lifecycle Benefits (pensions, health, education) Taxes (income, payroll) 1890s to 1990s Decadal Birth Cohorts Source Leg d Rel Birth Cohort (BC) 2.3 1.1 Earnings Group (EG) 20.3 16.5 BC x EG 8.9 5.2 EG x Sex 2.5 1.2 BC x EG x Sex 0.2 0.1 Individuals within Groups 65.8 75.8 (Wolfson and Rowe, 1997)
What About the Age of Retirement? Nowadays a major current source of structural strain is the long-term failure of our institutions to accommodate the steady rise in the proportion of people who are old. Large strata of older people have been added at the top of the traditional age pyramid, but no comparable activities have been prescribed for them either in the work force or the family; and no adjustments have been made for repercussions in all the other strata.. This "structural lag" means (apart from individual dislocations) that human resources in the oldest-and also the youngest-strata are underutilized, and excess burdens of care are imposed upon strata in the middle years. Matilda White Riley, Presidential Address, American Sociological Association 1986 (p9-10) 29
Alternate Views of the Aging Burden (LifePaths estimates) Demographic Ratios 0.50 0.45 0.40 0.35 0.30 0.25 Old Age Ratio Total Ratio 0.20 0.15 1 0.10 0.05 0.00 301971 1981 1991 2001 2011 Boomerangst 2021 Vancouver 2031 Feb 2011 23-Aug-11
Alternate Views of the Aging Burden (LifePaths estimates) Demographic Ratios 0.50 0.45 0.40 2 0.35 0.30 0.25 Old Age Ratio Total Ratio 0.20 0.15 1 0.10 0.05 0.00 31971 1981 1991 2001 2011 Boomerangst 2021 Vancouver 2031 Feb 2011 23-Aug-11
Alternate Views of the Aging Burden (LifePaths estimates) Demographic Ratios Annual Paid Hours of Work / Person 0.50 1000 3 0.45 Female 0.40 0.35 2 800 Male All 0.30 0.25 Old Age Ratio Total Ratio 600 0.20 400 0.15 1 0.10 200 0.05 0.00 0 321971 1981 1991 2001 2011 Boomerangst 2021 Vancouver 2031 Feb 1971 2011 1981 1991 2001 2011 23-Aug-11 2021 203
Alternate Views of the Aging Burden (LifePaths estimates) Demographic Ratios Annual Paid Hours of Work / Person 0.50 1000 3 0.45 Female 0.40 0.35 2 800 Male All 0.30 0.25 Old Age Ratio Total Ratio 600 0.20 400 0.15 1 0.10 200 0.05 0.00 0 331971 1981 1991 2001 2011 Boomerangst 2021 Vancouver 2031 Feb 1971 2011 1981 1991 2001 2011 23-Aug-11 2021 203
0.50 Alternate Views of the Aging Burden (LifePaths estimates) Demographic Ratios 1000 Annual Paid Hours of Work / Person 3 0.45 Female 0.40 0.35 2 800 4 Male All 0.30 0.25 Old Age Ratio Total Ratio 600 0.20 400 0.15 1 0.10 200 0.05 0.00 0 341971 1981 1991 2001 2011 Boomerangst 2021 Vancouver 2031 Feb 1971 2011 1981 1991 2001 2011 23-Aug-11 2021 203
Leisure Is Not Always the Most Valued Use of Time (2005 GSS) Age Groups all* males females 15-24 25-44 45-64 65+ Cleaning 2.4 2.2 2.6 2.1 2.4 2.5 2.8 Groceries 2.7 2.5 2.8 2.6 2.6 2.6 2.9 Maintenance 2.9 3.3 2.5 2.6 2.9 3.0 3.1 Other Shopping 3.0 2.5 3.4 3.5 2.9 2.7 2.9 Communting 3.0 3.0 3.0 2.8 3.0 3.2 3.8 Clubs 3.1 3.0 3.1 3.4 3.1 2.9 3.0 Volunteering 3.3 3.0 3.5 3.1 3.2 3.4 3.5 Cooking 3.3 3.1 3.4 3.1 3.3 3.3 3.3 TV 3.3 3.3 3.2 3.4 3.2 3.1 3.6 Social Events 3.5 3.3 3.7 3.8 3.5 3.3 3.2 Movies / Plays 3.7 3.7 3.7 4.3 3.9 3.4 2.9 Paid Work 3.8 3.8 3.8 3.7 3.7 3.9 4.2 Dining Out 4.0 3.8 4.1 4.0 4.1 3.9 3.7 Supper at Home 4.0 4.1 4.0 3.8 4.1 4.1 4.1 * basis for sorting five-point scale with 1 being dislike a great deal and 5 being enjoy a great deal 35
Use It or Lose It Rohweder and Willis, JEP 2010 36
Concluding Comments yes Houston, there are problems but they may not be the ones you re thinking of: retirement income adequacy, not only affordability phase in for C/QPP expansion more rapidly needed inter-generational equity yes, but do properly private savings unlikely to work Ambatscheer ideas (MER, maturing to annuities) good recall 1983 Frith Task Force re tax incentives retarget age of retirement definitely worth exploring for health and social as well as economic reasons