Assessment of the Amendment to the National Pension Act

As mentioned above, the amendment to the National Pension Act is highly likely to bring considerable change in the contribution-benefit structure of the National Pension Scheme as it contains a fairly large number of benefit cuts, an introduction to the Basic Old Age Pension System, and the rationalization of the related benefit level. To assess such effects of the revised act, a multifaceted review is necessary, one which covers the various aspects of both pension finance and the social security function. In this context, Section 3 shows the results of estimating the effects of the revision by analyzing the long-term sustainability of the National Pension Scheme, equity among different generations and income classes, adequacy of the benefit level, and responsiveness to changes in social environments.² In particular, in assessing the effects on each sector, various simulation methods have been used for quantitative analysis.

3.1 Sustainability

The purpose of the amendment to the National Pension Act is to obtain long-term financial stability of the National Pension Scheme in preparation for an aged society. Thus, it seems reasonable to prioritize the evaluation of how much the National Pension Act's amendment can contribute to the enhancement of the sustainability of the scheme. To analyze the impact of the amendment and to reflect the unique conditions of Korea's National Pension Scheme, this study used the long-term Actuarial Projection Model, a customized form of the World Bank's Pension Reform Options Software Toolkit (PROST) Model. Also, KNSO's population-related statistics were used as the basis for the analysis.³

In this actuarial projection, the most important economic variables are the assumptions over the wage growth rate, the interest rate, and the economic growth rate. The assumed values should be set up in full consideration of the dynamic relationship among these variables because their relationship plays a pivotal role in the assessment of the cash-flow projection as well as the actuarial evaluation. In this study, actual data were used to analyze the period between 1988 and 2007, while assumed values for real economic growth rate, real wage growth rate, and real interest rate were used for the future period (up to 2050). These values were calculated based on the Growth Accounting Method presented in the study by Han et al. (2006).⁴ Figure 2 [ PDF 77.5KB | 1 page ] presents the assumed values of such economic variables.

In long-term actuarial projections, it is necessary to differentiate the concepts of interest rate and rate of return on the pension fund to generate more accurate results. In fact, in a funded pension scheme, the pension fund's rate of return will have a direct impact on the pension finance, and the assumed value of the rate of return is not necessarily equal to the assumed interest rate value. That is, the return rate of the pension fund can be differentiated, depending on how efficient the fund management is or how strategically the assets are allocated. However, it is also difficult to make an assumption that there will be a constant gap between the return rate of the fund and the presumed interest rate. Therefore, this study has set an underlying assumption that the national pension fund's return rate is equal to the assumed value of the interest rate, following the universal rule. However, in the latter part of this study that covers policy discussions, different values were used to analyze the impact on the pension finance when the fund's rate of return is different from the assumed interest rate.

Under this assumption, Table 2 [ PDF 18.9KB | 1 page ] shows the results of comparing the financial prospects for the pension fund before and after the amendment of the National Pension Fund Act. As shown in Table 2, the instability of the pension finance is expected to be alleviated considerably in the long-term due to the second financial stabilization measures introduced in 2007. More specifically, with the stabilization measures implemented, the financial deficit of the pension fund is likely to occur in 2035, 10 years later than the previously expected point of time. Also, under the assumption that there will be no additional institutional changes, the point of time when the fund runs out of money will be postponed by 15 years from 2045 to 2060. Moreover, the ratio of fund reserve amount to gross domestic product (GDP) is likely to reach 46.2% by 2035, while the speed of decrease in the fund size is expected to slow down after the amendment.

Therefore, it is estimated that the amendment to the National Pension Act has considerably improved the sustainability of the pension finance through the decrease in benefit levels and significantly alleviated the structural imbalance between pension benefits and contributions at the same time. In particular, such a gradual decrease in the average income replacement rate based on insured people with 40 years of coverage by 20 percentage points (from 60% to 40%) is not only substantial, it is also meaningful that this adjustment has been applied at the initial stage of the system implementation. Nevertheless, the financial stabilization measures do not seem to be sufficient for fundamentally changing the “high benefit-low contribution” structure of the National Pension Scheme and achieving the actuarial balance of the pension fund. As shown in the projection results, the gap between pension benefits and contributions still remains despite a huge decrease in benefit expenditure in the long term (a 2.5 percentage point decrease against the projected GDP value in 2070).

As a result, it seems inevitable that the pension fund will be exhausted by 2060 if the current contribution rate stays at the same level. After that, the financial system will have to be replaced by the pay-as-you-go (PAYG) system. In this case, however, the contribution burden on the workers of future generations will become much heavier. As shown in Table 2, the per capita contribution rate of economically active people after the conversion to the PAYG system is expected to surge from current 9% to 23.9% by 2065, and to 24.3% by 2070. That is, compared to the situation where the National Pension Scheme continues to have the current partially funded system, not only will the future workers have to bear the burden of over 1.5 times higher contribution payments (refer to Section 4 for more information), but also they will receive benefits far less valuable than the present value of what they have paid, which will inevitably cause an excessive income transfer from next generations to current generations in the future.

To analyze the role of the National Pension Act's second amendment in alleviating the fundamental imbalance of the pension scheme and its limitations, it seems more reasonable to conduct the actuarial valuation of the National Pension Scheme using a “closed measure.” Table 3 [ PDF 22.2KB | 2 page ] and Figure 4 [ PDF 21.6KB | 1 page ] present the calculation results of the changes in the scale of total pension liability and total reserve for losses, using the “projected benefit obligation” (PBO) method. In Table 3, the “unfunded actuarial liability” (UAL) means the difference between total pension liability and actual fund reserve while the funding ratio is the ratio of fund reserve to the total pension liability. Therefore, in the case of using a fully funded pension scheme, the UAL value becomes zero (funding rate will be 1.0), while in the case of using the PAYG, the UAL value will be equal to the total reserve for loss.

As observed in Table 3, with the amendment to the National Pension Scheme, the size of UAL is expected to be reduced considerably from 278 trillion won in 2005 (pre-amendment period) to 150 trillion won (post-amendment period). The reason this UAL value decreases with the reform is because the PBO method considers the future subscription period of current pensioners in calculating the amount of total pension liability. That is, if the benefit rate for the future subscription period decreases with the amendment, the average payment rate over total subscription period will also decrease, which, in turn, will have an impact on the size of the past service liability. As a result, the funding rate is estimated to have increased from 38% before the amendment to 53.5% after the amendment, which will considerably improve the actuarial imbalance.

Despite such improvement, however, the ratio of the UAL to GDP is likely to increase continuously after 2030 and reach 100% around 2070. The funding ratio will also plunge after it reaches its peak rate—over 70%—by 2035. This implies that the actuarial imbalance of the National Pension Scheme still remains and it will actually be aggravated as the pension system enters a mature stage and population aging accelerates. Therefore, to strike a more fundamental financial balance for the National Pension Scheme, it seems necessary to make an upward adjustment of the current contribution level, or additional decrease in pension benefit within a short time frame.

The financial stabilization measures included in the revised bill for the National Pension Act, which was recently passed in the National Assembly, is different from what was originally submitted by the government in 2003, both in terms of pension benefit and contribution levels. Originally, the government planned to decrease the benefit level (average income replacement rate) from 60% to 50% and gradually increase the contribution rate from 9% to 15.9%, based on the results of the first actuarial valuation carried out in 2003. This conforms to the financial stabilization goal of keeping the reserve ratio of the pension fund at a level higher than 200% by 2070 (NPDC 2003).⁵ Despite some criticisms over the stabilization goal itself, the original plan can be positively viewed as progress, in that it set a specific goal for financial stabilization, unlike the pension reform in 1998, and attempted to fundamentally adjust the imbalanced benefit-contribution structure.⁶

However, core contents of the government's original revised bill were drastically changed during the four years of adjustment process at the National Assembly. That is, while the average income replacement rate was reduced to 40% in the long term to complement the adoption of the Basic Old Age Pension System, the contribution level was determined to be maintained at the existing level, considering the possible opposition and resistance from participants. However, because this adjustment was the result of political compromise without a full reflection of the specific financial goals for the long-term stabilization of the pension finance, the adjustment caused a repeat of the stopgap situation where measures were taken only to respond to the problems rather than fix the source of the problems, just like during the process for the first amendment to the National Pension Act in 1998. In this context, it seems an objective assessment to say that the long-term financial stabilization measures under the second amendment were only half successful.

3.2 Equity

To evaluate the effect of the amendment to the National Pension Act in terms of equity, it would be useful to differentiate equity among different income classes and the equity among different generations, and then analyze specifically how much the amendment has affected the value of the benefit by income class and by generation. In this study, the changes in the benefit values by generation and by income class before and after the amendment have been compared through an actuarial simulation method. Following are the assumptions and major variables used in the simulation analysis.

3.2.1 Simulation Procedures

In this study, sample subscribers were classified by age group and income class. By age group, single male workers at ages 55, 40, and 25 as of 2008 were selected while each income class was composed of pensioners with incomes of 1/3, 1/2, and 2 times the basic income level, respectively. The basic income level was calculated based on the average income of economically active people by each age group and the average income was calculated based on the data from the 2005 Basic Statistical Survey of Benefit Structure (Ministry of Labor 2005). More specifically, each average income of male workers at age of 52, 37, and 22 were set as 3,096,366 KRW; 2,940,189 KRW; and 1,180,269 KRW, respectively. Also, it was presumed that all the workers started to subscribe to the National Pension Scheme at age 25 and would continue to work until they reach the entitlement age and can receive pension benefits without any withdrawal from the subscription in the middle of the subscription period.

In this analysis, calculation of the pension benefits amounts after the retirement of the sample subscribers required not only their annual earnings level but also an assumption of the change in their income level as they grew older. That is, to draw up an “age-earnings profile” over the entire working period for each worker, additional assumptions are necessary for the “general rate of promotion” reflecting the increase in age and working period as well as the average wage growth rate. In this study, the general rate of promotion was calculated based on the “curve-fitting” value of the average income distribution in the statistical data of the Ministry of Labor. Then, the calculated rate has been equally applied to all age groups under the assumption that the distribution pattern will remain constant in the future. However, since the objects in this analysis were presumed to work continuously until they start to receive pension benefits, it has also been assumed that there will be no changes in the income level after they reach 50 in the virtual age-earning profile.⁷

Furthermore, it was necessary to adjust the contribution and benefit levels of each generation using the “death rate by age” as a weight, in order to estimate the size of the expected benefit value. However, due to limited data on the survival probability of each generation for this study, the benefit value was calculated based on the expected years of life, presuming that each pensioner lives until 55, which is the entitlement age for the current Early Old Age Pension. The expected years of life were calculated based on the KNSO's 2005 Complete Life-Table (by generation). The calculation result shows that the life expectancy of male workers aged 55 as of 2008 is 78.5, while those aged 40 and 25 as of 2008 are 80.5 and 82.5, respectively, revealing a continuous increase in the pensionable service period over time.

As for the institutional variables, the effect of the amendment to the National Pension Act has been analyzed through a comparison between the periods before and after the amendment in July 2007. For this analysis, it has also been reflected that the starting age eligible to receive the benefit of the Basic Old Age Pension will gradually go up from the current 60 to 65 by 2033. In calculating the pension benefit level by each pensioner, the function of income redistribution of the national pension should be considered. That is, because the current scheme is composed equally of the “earnings-related portion” which is based on the recalculated average life-time earnings (Value B) of individual pensioners and the “fixed portion” estimated based on the average earnings of the last three years during the pensionable service period (Value A), an assumption for calculating Value A is also necessary. Therefore, it has been assumed in this study that Value A increases in accordance with the real wage growth rate of each year from 1,412,428 KRW, the actual average earnings value in 2005.

In the application of contribution value, it has been assumed that workers pay the total amount of the contribution because the employers' share of the contribution payment will ultimately be transferred to workers through a decrease in the amount of real wage. The contribution rate has been assumed to be 3% by 1992 after the introduction of the National Pension Scheme, 6% between 1993 and 1997, and 9% after 1998. Also, as for the values of real wage growth rate and real interest rate, which are playing a critical role in the actuarial profitability assessment, the same values used in the previous actuarial projection have been used in this analysis as well. Finally, all the price variables used in this analysis are based on the constant price in 2005.

3.2.2 Results

Table 4 [ PDF 20.6KB | 2 page ] demonstrates the major indices that show the changes in benefit values before and after the amendment to the National Pension Act, analyzed through a simulation under the above assumptions and conditions. The equity effect of the amendment could be analyzed by calculating the difference in the net value of pension benefits by each age group and income class between pre- and post-implementation periods. Table 4 presents the internal rate of return (IRR) and “benefit-to-contribution ratio”; i.e., the ratio of the present value of the expected pension benefits to total contribution payment, which is the basis of the calculation.

3.2.2.1 Intragenerational Equity

First, in terms of the benefit-to-contribution ratio, there is a big difference in the net pension values of the old National Pension Scheme among different income classes, reflecting the income redistribution effect of the “fixed portion” of the scheme. For example, in the case of workers aged 55 years as of 2008, the total benefit of average income earners is expected to be about twice the amount of their total contribution payment for 21 years based on the present value, implying that the existing National Pension Scheme has a serious imbalance between “high benefit and low contribution.” Meanwhile, the benefit-to-contribution ratio of the workers with 1/3 of the average income level is 3.4, meaning that the lower the income, the bigger the gap between the present values of contribution and benefit.⁸

In Table 4, the benefit-to-contribution ratio of the workers who earn twice as much as average workers is not much lower than that of the average income earners, because there is a ceiling on the pensionable income (currently 3.6 million KRW). Nevertheless, in the case of the workers aged 55 as of 2008, the first generation of the National Pension Scheme, the benefit-to-contribution ratio for the high-income class is about 1.9, much higher than the actuarial balance level, where the benefit-to-contribution ratio is 1.0. This structural imbalance is also revealed in the calculation result of internal rate of return. For example, in the case of the 55-year-old age group, the real IRR of the income class with one-half of the average income level, which is similar to middle-income subscribers in the National Pension Scheme, is estimated to be over 9%. This is more than twice the level of the real interest rate or the real rate of return on the Pension Fund, meaning that the benefit level for the early generation of the scheme is relatively too high in terms of actuarial value.

In this second amendment to the National Pension Act, as a downward adjustment has been uniformly applied to all benefit levels, the cutting ratios in different income classes are almost proportional. Therefore, it seems possible to assess the impact of the amendment on the equity among different income classes as “relatively neutral.” Such proportional effect of the benefit cut is also observed in the 40-year-old age group and 25-year-old age group. However, considering the fact that the newly introduced Basic Old Age Pension benefit will be paid to elderly people who fall in the lowest six income deciles, the overall income redistribution capacity of the National Pension Scheme could be strengthened further.

Despite the benefit cuts, the benefit-to-contribution ratio of workers aged 55 is still very high in every income class. This is so because, in the context of protecting the vested interests of existing pensioners, the benefit level under the old act will be guaranteed for the previous subscription period and the decrease in pension benefits will occur gradually over a long period of time. Therefore, in the case of workers aged 55 as of 2008, the benefit-tocontribution ratio is likely to fall only by about 4% in all income classes. However, the decrease in benefit-to-contribution ratios is expected to grow bigger for the future generations.

3.2.2.2 Intergenerational Equity

The gap of net pension benefit values among different generations under the National Pension Scheme and the effect of the amendment to the National Pension Act could be assessed through a comparison of benefit-to-contribution ratios and IRR among different age groups. Table 4 shows the benefit-to-contribution ratios and IRR of three age groups: 55, 40, and 25. Under the existing pension scheme, these two values seem to decrease further with future generations due to the first reform measure to cut the benefits in 1998. However, as seen in Table 4, the benefit-to-cost ratio of the national pension is above 1.0 in every age group and income class, implying that the net pension benefit is applied to all of the current pensioners. In other words, the old National Pension Scheme contained a very serious problem of a structural imbalance.

The second amendment to the National Pension Act is expected to cause a considerable change in the benefit level by each age group. As presented in Table 4, the extent of decrease in the benefit-to-contribution ratio of workers aged 55 as of 2008 is only 4% for each income class, while that of workers aged 40 and 25 is about 16% and 29%, respectively. As a result, the gap of benefit-to-contribution ratios between average income earners aged 55 and those aged 25 increased from 17.2% (1.98 and 1.64 for age group 55 and 25, respectively) before the amendment to 38.8% (1.90 and 1.16 for age group 55 and 25, respectively) after the amendment. This is, of course, due to the effect of benefit cuts under the amendment occurring slowly, which in turn, is the result of grandfathering over the past subscription period and a gradual decrease in the level of future benefits.

As noted above, the study suggests that the gap of net pension benefits among different age groups gets wider with the amendment to the National Pension Act. However, this should not be interpreted that the amendment has a negative impact on intergenerational equity. As presented above, under the existing scheme, the conversion to PAYG system will inevitably increase the contribution rates of future generations to over 30%. However, it is important to note that the second amendment to the act cuts the net pension benefits of early generations, thereby lessening the burden of contribution payment for future generations. That is, since such a result inevitably occurs during the accrual period due to the grandfathering measure, from a long-term perspective, the amendment to the National Pension Act will have to be assessed to have enhanced the intergenerational equity by alleviating the structural imbalance of the National Pension Scheme.

Meanwhile, the benefit-to-contribution ratios of all age groups and income classes exceed the system balance level (1.0) even after the amendment to the National Pension Act, which indicates that all of the current pensioners are likely to receive considerable net pension benefits. Of course, there is a possibility that the benefit-to-contribution ratios of some highincome earners will fall below 1.0, in case the objects of actuarial projection include more generations in the future. However, even considering this, the structural imbalance of the National Pension Scheme will continue to exist after the amendment. Therefore, additional measures to rectify remaining structural imbalance should be followed to improve intergenerational equity. In particular, additional increase in pension contribution should be promptly applied for the enhancement of financial stabilization and equity among different generations.

3.2.2.3 Adequacy

Adequacy is an assessment standard used to estimate whether the revised National Pension Scheme will play a sufficient role in terms of old age income security. However, the adequacy of such old age income replacement level should be judged in consideration not only of the single-layered National Pension Scheme but also of the multilayered other pension plans including retirement pension at the company level, private pension at the individual level, and savings for old age. Also, the replacement level of proper old age income will differ depending on the type of family, health status, or whether or not there are any other income sources for the pensioners. Therefore, it is difficult to suggest an absolute assessment standard to evaluate the adequacy of the National Pension Scheme. The adequacy level should be assessed indirectly by calculating how the income replacement rate of the National Pension Scheme has changed before and after the amendment.

Figure 3 [ PDF 19.8KB | 2 page ] shows the extent of changes in the income replacement rate of the pension benefit against the reestimated average lifetime earning during the working period of each pensioner in accordance with the amendment to the National Pension Act. As presented in Figure 3, the scale of decrease in the income replacement rate caused by the amendment is small for workers aged 55 who have a relatively longer subscription period, and the extent of decrease grows bigger with future generations. In particular, the income replacement rate of those aged 25 decreases by almost 30% in every income class after the amendment to the Act, which suggests that its function of serving as an old age income source will be weakened over time.

Of course, the old age income security level of the National Pension Scheme differentiates among each income class. In the case of the y/2 group, which is the equivalent of the medium-income subscribers of the National Pension, the income replacement rate is expected to remain over 40% in the long term even after the amendment to the National Pension Act, and it is likely to play an important role of social insurance. Also, its effect of old age income security is clearer for the low-income class (y/3), due to its income redistribution function. However, in the case of average income earners, the income replacement rate after the amendment will remain a little above 30% in the long term, even though they work during the entire subscription period. In particular, the income replacement rate of highincome earners (2y) seems to decrease to around 17% in the long term.

That is, although the pension fund serves as an old age income source to a certain extent despite the cut in pension benefits, its capacity to provide old age income security seems to be weakened for average and higher income earners. Therefore, to complement such a function, it will be necessary to discover various sources of old age income for the mid-tohigh- income classes.

The remarkable decrease in the income replacement rate of high-income earners (2y) in Figure 3 is due to the application of the earnings ceiling for the pensionable income and the income redistribution effect of the National Pension Scheme. That is, under the current scheme, earnings exceeding 3.6 million KRW will be excluded from the objects of both contributions and benefits, and therefore, the replacement rate of high income earners, whose income levels are higher than the earnings ceiling, will decrease further. However, despite the increase in the real income and price level, the upper earnings limit has been fixed at its nominal value (3.6 million KRW) for the past 20 years, since the National Pension Scheme was first introduced in 1988. For this reason, the present value of the current earnings ceiling is about 120% of the average income level, resulting in restrictions on additional opportunities for applying the national pension scheme to a large number of midto- high income earners. In particular, since the role of old age income security for relatively higher income earners is likely to be weakened with the benefit cuts, it seems inevitable to readjust the earnings ceiling to a more reasonable level and expand the function of social insurance.

Download this Paper [ PDF 228.1KB| 29 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.