As there is also a drive for private pensions in this country, I have been doing some re-calculating on pensions, with dutch income data since 1970 (
www.gemiddeld-inkomen.nl/inkomens-vanaf-1970.php).
I'm taking a person who started working at age 26, and starts his pension at 67. He's been saving 20% for his pension for all 42 years, and he has received 2% interest on these savings annually. That means he ends up with € 243,297.47 in savings.
The average 26-year old is expected to live until 81. So that is 14 years of pension. If he takes 57.3% of the average wage in pension, his fund will be depleted at exactly that age. If this pension has an insurance structure, he will probably be able to receive a pension beyond that age from people who died before 81.
I don't know if 57.3% is realistic... He doesn't need to pay for his pension, which was 20%. Leaves a 20% drop. Is that too much?
Assuming it is reasonable, though, how does it compare to direct pension funding (i.e. working people pay directly for the pensions of retired people)? On average, every worker turns into a pensioner for 14 years (in my model). And he works 42 years, which is 3 times the length of his pension, which means there are 3 times more workers than pensioners. A 20% pension tax will therefore be able to sustain retired folk at 60% of income: slightly better than saving.
If we say 57.3% is enough for them, you have (60%/57.3%=) 4.7% extra funds. That means every 21 workers can sustain 22 pensioners. Put differently, at any time you can have 4.7% more pensioners than workers without running a deficit in the pension funding. The baby boom generation is probably bigger than that, but it will pass. If the fund requires borrowing for the baby boom generation, it will be able to repay those loans at some later point with the extra funds.
It surprised me that this time around, the 2 alternative systems perform comparably. The primary reason for this is the fact that I used a different income growth calculation. Over the last 40 years, income in this country has not grown by a percentage, but more linear. So I used a linear formula to extrapolate the expected income beyond 2011. Note that this means that when inflation and economic growth are percentages, a linear growth of salaries means the average worker effectively earns less and less.
The faster salaries grow (for example if they were to be linked to economic growth), the more it starts to look like my earlier calculations, and thus the less favourable it is to the savings model. Conversely, we have to conclude that when economic growth and inflation levels off, and salaries do not rise as much any more, the savings model actually becomes more interesting than the alternative of direct pay - though only if the fund continues to pay interest on savings.
However, since you can't retro-actively save money, I don't see how that solves the demographic time bomb problem at all.
