All entries for Thursday 09 April 2015

USS reforms: DC contributions in the hybrid scheme will be too little too late

Mike Otsuka of LSE has used the USS online modeller to make some interesting comparisons to show how much worst value for money the new DC (defined contribution) section will be compared with the DB (defined benefit - CRB or FS).

I've now posted a follow up to my "Using USS's modeller to show how much worse DC is than DB" post. This one is called "DC contributions in the hybrid scheme will be too little too late". Here it is on Facebook:

https://www.facebook.com/mike.otsuka.9/posts/350697368463862:0

The text and a chart are also below my signature.

Best,
Mike

personal.lse.ac.uk/OTSUKAM/

DC contributions in the hybrid scheme will be too little too late:

This chart illustrates a fairly serious design flaw in the proposed hybrid DB/DC scheme. It compares the new CRB defined benefit (DB) pension that a £55,000 salary will generate with the equivalent annuity that one could purchase if one earns £55,000 in the new defined contribution (DC) section.

What the chart shows is that one needs to keep one's DC pension pot invested for a long time, and at a fairly high rate of return, in order to generate enough wealth to purchase an annuity that matches or exceeds the CRB DB pension. Even at the USS consultation modeller's highest rate of return of 7.5%, one needs to keep one's DC pension pot invested for over 30 years in order to purchase a matching annuity.

Academics in USS won't, however, start putting much money into their DC pension pots until they're earning higher salaries later in their careers. Perversely, members will end up making more and more DC contributions when they will yield less and less of a pension per pound contributed. This flaw in the system will be magnified if the default DC fund involves the common practice of 'lifestyle' de-risking of one's pension pot into lower-risk, lower-return assets as one nears retirement.

This problem would be mitigated somewhat if USS members are allowed to keep their DC pension pots invested in USS funds throughout their retirement. Under this scenario, pension income would be generated via a gradual drawing down of their pot in retirement, rather than via an exchange of their entire pot for an annuity at the beginning of their retirement.

A move to collective DC would provide further mitigation. Under CDC, USS could make use of investment and longevity risk pooling, plus a continual and long-term investment of the collective fund in return-seeking assets, in order to try to achieve the target of a pension that is equivalent to the CRB DB pension.

An illustration of how much worse the defined contribution (DC) part of the proposed USS hybrid pension is than the defined benefit (DB) part:

Assume that you are a man who will turn 45 in 2016 and who will retire twenty years later at 65. According to the USS consultation modeller, if you earn exactly at the DB/DC threshold of £55,000 + CPI for those 20 years, that will generate a DB pension of £14,310 per year, plus a lump sum of 3 times that amount, in today's pounds. The cost of this pension to you is an employee contribution of 8% of your gross salary.

If, at the same 1/75 accrual rate, the threshold had been twice as high (£110,000 + CPI), and you had earned at precisely that higher threshold for 20 years, then your pension would have been twice as great: £28,620 per year in today's pounds.

Suppose, now, that you earn twice the DB/DC threshold under the current hybrid DB/DC proposal. How much and how well, according to USS's modeller, would you need to invest your DC pot tied to the £55,000 of your earnings about the threshold in order to match the pension of £14,310 per year (plus a lump sum of 3 times that amount) on the DB portion of your pension below the threshold?

To answer this question, let us first assume no extra employee contributions above the required 8%. Let's also assume the default setting of 5% annual returns on one's DC pension pot. Here the modeller indicates that the annuity (plus lump sum of 3 times that amount) that you could purchase would be worth only £7,060 per year, or less than half of what the DB pension provides for the first £55,000 of your income. Even if you slide the modeller setting all the way up to the highest returns of 7.5% per year, that would generate a pension of only £9,687 per year. At that high rate of return, you would need to contribute an additional 3% of your gross income (or 11% total) to accumulate a pension pot large enough to purchase an annuity equivalent to the DB pension of £14,310 on the first £55,000. If we assume USS's default return of 5% per year, your employee contribution would need to be twice as great as for your DB pension (i.e., 16% rather than 8%) in order to generate an equivalent pension.

A 5% return is twice USS's assumed rate of inflation of 2.5%. There is no investment available that guarantees such a return. Perhaps it would be possible to invest in inflation-linked bonds that are nearly guaranteed to keep pace with inflation. But one would need to spend 21% of one's income (13% in addition to one's 8% employee contribution) to match the £14,310 DB pension if one's rate of return is at the assumed 2.5% rate of inflation.

Dennis Leech : 09 Apr 2015 12:43 |  Comments (0) |  Close comments |  Report a problem