The distributional effects of the pension system reform in Poland

	Pre-reform	Post-reform
Type of system	Pay-as-you-go defined benefit	Pay-as-you-go, NDC plan (first pillar) + FDC plan (second pillar)^*
Benefit formula	Flat rate plus a component based on earnings and tenure	Depends on contributions paid on lifetime earnings and LE at the retirement age
Transition from old to new system	Cohorts born before 1949 – covered by the old system Cohorts born between 1949 and 1968 could choose to participate only in NDC or in both NDC and FDC pillars Cohorts born after 1969 fully covered by the new system

FDC, funded defined contribution; LE, life expectancy; NDC, notionally defined contribution.

* In 2013, a part of the contributions paid into the second pillar was moved to the first pillar and indexed by an average GDP growth from 5 years before indexation. Furthermore, the reform of 2013 established that the first pillar, ZUS, will handle the pension funds retirement plans, with the accumulated funds transferred incrementally 10 years before the statutory retirement age. Since 2014, the NDC scheme is the default option; the insured can opt-out to allocate part of their contributions to the FDC scheme. If the insured decides to pay their entire contributions to the NDC scheme, they will be indexed as pension capital in ZUS.

Note: The information is according to the legal situation in December 2018.

Source: OECD (2017), ZUS (2017), and own elaboration of authors.

Table 2

Sample descriptive statistics

Variable	Sample 1			Sample 2
Variable	Mean	Minimum	Maximum	Mean	Minimum	Maximum
Dependent variable
Log monthly earnings	7.54	5.70	8.70	7.66	5.70	8.70
Characteristics of individuals
Tenure	20.60	0	65	17.73	8	28
Educational attainment (%)
Incomplete primary, primary and gymnasium	7.30			5.57
Basic vocational	39.71			40.83
Upper secondary general	7.24			5.90
Upper secondary vocational	24.82			24.12
Postsecondary non-tertiary	1.48			1.37
Tertiary	19.44			22.22
Occupation (%)
Public servants and managers	4.22			5.18
Scientists and researchers	10.73			12.78
IT and technical personnel	8.34			9.29
Administration	5.29			5.09
Service workers	8.59			7.48
Manual workers and workers in construction	31.73			31.32
Other types of manual workers	21.29			22.23
Elementary occupations	9.10			6.64
Industry
Mining	3.33			4.05
Manufacturing	29.14			30.46
Electric, gas and sanitary service	3.61			2.92
Construction	18.19			17.16
Trade	11.76			12.59
Transportation, communications, information	13.41			13.32
Finance, insurance, and real estate	2.37			2.39
Services	6.22			4.14
Public administration, education, and health	11.98			12.98
Sample size, observations	15,940			4,518

[i] Note: Sample 1 is the estimation sample for earnings profiles; Sample 2 is the sample of working men born between 1969 and 1979 for pension benefits calculation.

The predicted age-earnings profiles for men aged 18–64 years by education level, in euros.
*Notes*: Earnings are net monthly earnings in EUR 2012. Own calculations based on PHBS, 2012. Three groups of education levels include the following groups from Table 2: Group 1: Incomplete primary, primary, and gymnasium; Group 2: Basic vocational, upper secondary general, upper secondary vocational, and postsecondary non-tertiary; Group 3: Tertiary. PHBS, Polish Household Budget Survey.

Table 3

Distribution of net monthly pension benefits at the age of 65 years in euro, 2034–2044

Year of birth	N	p90/p10	p90/p50	p10/p50	Gini	p90/p10	p90/p50	p10/p50	Gini
Year of birth	N	New pension system (baseline)				Old pension system (counterfactual)
1969	378	2.72	1.69	0.62	0.219	2.21	1.61	0.73	0.178
1970	360	2.85	1.71	0.60	0.220	2.16	1.56	0.72	0.172
1971	390	2.80	1.71	0.61	0.227	2.24	1.60	0.71	0.182
1972	388	2.88	1.73	0.60	0.228	2.27	1.61	0.71	0.180
1973	394	2.82	1.79	0.64	0.236	2.21	1.65	0.75	0.185
1974	406	2.98	1.79	0.60	0.239	2.28	1.65	0.73	0.182
1975	417	3.03	1.90	0.63	0.248	2.28	1.68	0.74	0.187
1976	445	3.24	1.88	0.58	0.250	2.26	1.60	0.71	0.185
1977	412	3.27	2.04	0.62	0.259	2.35	1.74	0.74	0.186
1978	456	3.56	1.94	0.54	0.264	2.36	1.65	0.70	0.188
1979	472	3.27	1.81	0.55	0.267	2.21	1.55	0.70	0.187

[i] Note: Depending on the birth year, projections of pension benefits are for 2034 (birth year 1969) up to 2044 (birth year 1979).

Distributions of predicted monthly pension benefits at the age of 65 years, baseline scenario (post-reform) and counterfactual (no-reform) scenario, in euro.
*Note*: Exchange rate 2012: 1 EURO = 4.185 PLN.

Projected changes in p90/p50 and p90/p10 percentile ratios in the old and the new pension systems for men born in years 1969–1979.

Table 4

RRs for first pension benefits received at the age of 65 years

Year of birth	N	p90/p10	p90/p50	p10/p50	Average (%)	p90/p10	p90/p50	p10/p50	Average (%)
Year of birth	N	New pension system (baseline)				Old pension system (counterfactual)
1969	378	1.33	1.14	0.85	42	1.56	1.26	0.81	80
1970	360	1.33	1.13	0.85	42	1.62	1.33	0.82	82
1971	390	1.30	1.13	0.87	40	1.57	1.30	0.83	80
1972	388	1.30	1.13	0.87	39	1.57	1.31	0.83	78
1973	394	1.28	1.13	0.88	38	1.58	1.32	0.83	80
1974	406	1.27	1.12	0.88	38	1.60	1.32	0.82	81
1975	417	1.24	1.13	0.91	37	1.56	1.28	0.82	80
1976	445	1.21	1.12	0.93	36	1.58	1.32	0.84	80
1977	412	1.22	1.11	0.91	36	1.55	1.27	0.82	79
1978	456	1.15	1.08	0.94	35	1.57	1.30	0.83	78
1979	472	1.14	1.08	0.95	35	1.51	1.28	0.85	78

[i] RR, replacement rate.

[ii] Notes: Depending on the birth year, projections of pension benefits are for 2034 (birth year 1969) up to 2044 (birth year 1979). RRs are calculated as a ratio of the first pension benefits to last earnings before retirement.

Table 5

Impact of the minimum pension benefit on inequality of pension benefits and RRs

Scenario	Cohorts
Scenario	1969	1974	1979
	Gini index of pension benefits
Baseline – 65 years, new system	0.219	0.239	0.267
Counterfactual – 65 years, old system^*	0.178	0.182	0.187
New 65 years + min pension H	0.167	0.158	0.179
New 65 years + min pension L	0.199	0.210	0.241
	First pension benefits p90/p10 ratio
Baseline – 65 years, new system	2.72	3.03	3.27
Counterfactual – 65 years, old system	2.21	2.28	2.21
New 65 years + min pension H	1.99	1.89	1.97
New 65 years + min pension L	2.45	2.51	2.84
	Average RRs (%)
Baseline – 65 years, new system	42	38	35
Counterfactual – 65 years, old system	80	81	78
New 65 years + min pension H	48	47	44
New 65 years + min pension L	45	41	38

RR, replacement rate.

* Results for the counterfactual scenario with minimum pension are not reported. The impact of the minimum pension in the old system on inequality in pension benefits is almost negligible due to a generous redistributive formula. In addition, men in our sample have relatively long tenure. As a result, only <1% of pensioners under old rules would receive a minimum pension.

Table 6

Impact of higher retirement age and no difference in employment rate on pension benefits and RRs

Scenario	Cohorts
Scenario	1969	1974	1979
	Gini index of pension benefits
Baseline – 65 years, new system	0.219	0.239	0.267
Counterfactual – 65 years, old system	0.178	0.182	0.187
Higher retirement age (67 years)	0.221	0.241	0.268
No difference in employment rate	0.203	0.220	0.240
	First pension benefits p90/p10 ratio
Baseline – 65 years, new system	2.72	3.03	3.27
Counterfactual – 65 years, old system	2.21	2.28	2.21
Higher retirement age (67 years)	2.76	3.01	3.28
No difference in employment rate	2.58	2.84	2.95
	Average RRs (%)
Baseline – 65 years, new system	42	38	35
Counterfactual – 65 years, old system	80	81	78
Higher retirement age (67 years)	49	44	41
No difference in employment rate	47	43	40

[i] RR, replacement rate.

Table 7

Gini index of pension benefits, first pension benefits p90/p10 ratio, and average RRs for selected cohorts

Scenario	Cohorts
Scenario	1969	1974	1979
	Gini index of pension benefits
Baseline – 65 years, new system	0.219	0.239	0.267
Counterfactual – 65 years, old system	0.178	0.182	0.187
Higher LE	0.219	0.239	0.267
Higher rate of return	0.216	0.236	0.264
	First pension benefits p90/p10 ratio
Baseline – 65 years, new system	2.72	3.03	3.27
Counterfactual – 65 years, old system	2.21	2.28	2.21
Higher LE	2.72	3.03	3.27
Higher rate of return	2.70	2.94	3.27
	Average RRs (%)
Baseline – 65 years, new system	42	38	35
Counterfactual – 65 years, old system	80	81	78
Higher LE	40	36	33
Higher rate of return	48	44	42

[i] LE, life expectancy; RR, replacement rate.

Table A1

Assumptions for sensitivity analysis – various scenarios

	Baseline: new system	Counterfactual: old system	New system, with higher (H) minimum pension	New system, higher retirement age	New system, higher LE	New system, higher rate of return	New system, same employment rates for all education groups	Earnings not falling
Employment rates (%)	Higher education – 92.9, secondary education – 79.1, primary education and lower – 53.1						92.9 for all
Retirement age (years)	65			67	65
LE at retirement (in months) at the year of retirement	From 252 to 271	-	From 252 to 271	From 236 to 254	LE from the baseline scenario + 12 months (from 264 months to 283 months)	From 252 to 271
Rate of return since 2018 (%)	2	-	2			3
Wage profiles	Age-wage profiles rising until the age of maximum wage and then falling							Age-wage profiles constant since the age of maximum wage