Defined Contribution Vs Defined Benefit (USS Scheme)

Sam_J

I am a recent member of the USS pension for University academics. The terms of the defined benefit scheme are that you build up 1/75 of each years salary. Contributions are 8% of salary plus the employer pays 18% of salary. In addition, the employer will match a 1% contribution into a defined contribution scheme.

Initially I was very pleased to be part of this scheme. I anticipate working for around 30 years, so should build up 30/75 of my average salary, plus any contributions to the defined contribution scheme. However, upon learning of the significant deficit of the USS scheme I have become rather less pleased - it seems there are liabilities of around £77.5bn and assets of around £60bn, so a substantial deficit that is over a quarter of the scheme's assets. Most of this deficit is due to the liabilities from previous (more generous) final salary terms of the USS, that closed in late 2016. New terms were negotiated of average rather than final salary, 1/75s rather than 1/60s and an increase in employee and employer contributions.

So it seems that a decent chunk of the 8% + 18% that is going into "my" pension is in reality going towards paying the pension of members in the more generous final salary scheme, to which I have no access. In light of this, I have done a few calculations of the benefits of the defined benefit scheme vs investing all contributions into a defined contribution scheme.

Assumptions are that I work 30 years, and my salary increases match inflation - meaning the figures below are my inflation adjusted average salary. I contribute 9% of my salary and Employer contributes 19%. On the DC pot, I do 4% drawdown upon retirement.

Defined benefit: I end with around 42/43/45% of my average salary depending on market performance of the DC part of the scheme (market beats inflation by 0/2/4% respectively).

Defined Contribution: Various scenarios exist, depending on investment return and inflation. In each case I take the same 1.2 of average salary tax free lump sum from the pot upon retirement that I would get in the DB scheme, so the pots below are after this deduction.

Investment returns match inflation: Pot of 6.9 * average salary giving an income of 28% average salary with 4% drawdown.
Investment returns exceed inflation by 2%: Pot of 10.2 * average salary giving an income of 39% average salary with 4% drawdown.
Investment returns exceed inflation by 4%: Pot of 13.9 * average salary giving an income of 56% average salary with 4% drawdown.

The "breakeven" point for income based on that level of drawdown appears to be just over 2% above inflation returns, which over a 30 year period appears historically likely. But crucially, I will have the entire pot as an asset to increase income in scenarios where markets have performed poorly as I wish, or to pass on to children etc., which I do not have in the DB scenario. I will lose the payout to my spouse upon my death that I would get in the DB scheme, but this is mitigated by the fact that death is unlikely in the near future, I have life insurance, and my spouse can keep the pot upon my death in a DC scheme.

What do people think? The reason I ask is that the USS scheme is currently under review and there are talks about moving to a fully DC scheme. The union is likely to vote for a strike on this matter in opposition. However, I think for members who have only just joined the scheme that it would actually be beneficial to support a move to a fully DC scheme, provided employer contributions remain similar. That way we are not funding the final salary pensions of more senior academics, rather than our own, and I think will likely result in a more comfortable retirement for recent joiners to the scheme. If my thinking is correct on this matter then I may contact the union on that matter.

I am far from an expert in these matters, and would love a few thoughts.

kidmugsy

The DB and DC sections carry quite different risks: the risk of foundering vs the risk of your investing being unsuccessful.

You could always split the risks by carrying on in the DB scheme for a while, swapping to the DC scheme when you choose, or perhaps when you have no choice.

As a secondary consideration, bear in mind that DB members are often offered unusually good temporary terms when they are obliged to swap to DC.

Sam_J

kidmugsy wrote: »

The DB and DC sections carry quite different risks: the risk of foundering vs the risk of your investing being unsuccessful.

You could always split the risks by carrying on in the DB scheme for a while, swapping to the DC scheme when you choose, or perhaps when you have no choice.

As a secondary consideration, bear in mind that DB members are often offered unusually good temporary terms when they are obliged to swap to DC.

Yes, the reduced risks in a DB are definitely worth considering. That is why I tried to model a few scenarios of returns over the next 30 years. It would seem to me that I would have to be extremely unlucky to be worse off with 27% contributions going into a DC rather than DB scheme. I guess that is the question I am mostly asking!

In answer to the question in the second paragraph - I do not have a choice of moving to a DC scheme currently (only for contributions above the 8%, of which only the first 1% are matched by the employer). Ideally what I would like is for the union to ask the USS to give members of a choice between staying in the DB or moving to a DC with similar employer contribution. I would be sorely tempted to move to the DC, because a decent chunk of my pension contributions are currently going to fund someone else's pension. Am I thinking about this the wrong way?

(I currently have very little built up in this scheme, so the transfer valuation is not so relevant to y particular case)

Southend1

USS final salary was 1/80, not 1/60.

The scheme is 83% funded as at 31 March. Bear in mind historic low gilt yields are depressing this figure. The employer covenant is considered strong.

I believe 2% of the current employer contribution level is earmarked for past service deficit. Bear in mind the employer contribution rate is irrelevant for individual members who receive a defined benefit regardless of investment performance.

If USS switched to a D.C. only scheme, it is highly unlikely that employer contributions would remain at current levels - currently only 12% of salary is paid above the salary threshold into the DC scheme as employer contribution.

D.C. schemes are very high risk for individual members who don’t benefit from pooled investments and the economy of scale and security of the defined benefit.

I will be very cross indeed if the USS switches to D.C. only. It is a huge benefit to employees to have such a good defined benefit pension and one of the main reasons I continue to work for a university.

Sam_J

Southend1 wrote: »

I believe 2% of the current employer contribution level is earmarked for past service deficit. Bear in mind the employer contribution rate is irrelevant for individual members who receive a defined benefit regardless of investment performance.

This is useful information, thank you. Do you know if this is in a document somewhere I could read?

Southend1 wrote: »

If USS switched to a D.C. only scheme, it is highly unlikely that employer contributions would remain at current levels - currently only 12% of salary is paid above the salary threshold into the DC scheme as employer contribution.

Presumably the employer contribution would then be 17% for a 9% employee contribution based on what you said previously about only 2% going to fund the deficit.

Southend1 wrote: »

D.C. schemes are very high risk for individual members who don’t benefit from pooled investments and the economy of scale and security of the defined benefit.

I am aware of the additional risk, which is why I modelled various investment return scenarios over the 30 years I will likely work. It seems highly unlikely to me that I would be worse off with current contributions going into a DC scheme, even if market returns stay equal to inflation for the next 30 years (highly unlikely!)

Southend1 wrote: »

I will be very cross indeed if the USS switches to D.C. only. It is a huge benefit to employees to have such a good defined benefit pension and one of the main reasons I continue to work for a university.

Would you be happy for new members like me who might wish to make an informed decision to move to a DC scheme with 9%/17% contributions, to be given that choice?

Southend1

Sam_J wrote: »

This is useful information, thank you. Do you know if this is in a document somewhere I could read?


Presumably the employer contribution would then be 17% for a 9% employee contribution based on what you said previously about only 2% going to fund the deficit.



I am aware of the additional risk, which is why I modelled various investment return scenarios over the 30 years I will likely work. It seems highly unlikely to me that I would be worse off with current contributions going into a DC scheme, even if market returns stay equal to inflation for the next 30 years (highly unlikely!)



Would you be happy for new members like me who might wish to make an informed decision to move to a DC scheme with 9%/17% contributions, to be given that choice?

Sorry, it’s 2.5%, not 2%. See https://www.uss.co.uk/~/media/document-libraries/uss/how-uss-is-run/2015scheduleofcontributions.pdf.

Employers would be unlikely to pay anywhere near 17% to a D.C. only scheme. Currently it is 12% above the salary threshold. I believe this is an indication of where it would likely be initially if USS was switched to D.C. only.

Your appetite for risk is yours alone, but i can not afford to take that level of risk with my retirement funds. I am relying on USS for the bulk of my income in retirement, other than the state pension. A few years of poor returns pre retirement could mean I have to work many extra years or struggle. Pooled investments also benefit hugely from economies of scale - individuals in D.C. funds generally switch to low risk investments in the ten years or so before retirement- this protects from investment risk to an extent but also wipes out potential for returns that could be earned in a pooled D.C. or DB scheme.

I would not be happy for members to have a choice of a D.C. only option. Not only would it mean many of those who choose that option could well be worse off in retirement, it would undermine the employer covenant for the DB section and hasten its closure, making all of us potentially worse off in retirement.

DB pensions are extremely good value for individual members, hence why they are so expensive for employers. I would not want to give up this valuable benefit of working for a university.

Sam_J

Southend1 wrote: »

Sorry, it’s 2.5%, not 2%. See https://www.uss.co.uk/~/media/document-libraries/uss/how-uss-is-run/2015scheduleofcontributions.pdf.

Employers would be unlikely to pay anywhere near 17% to a D.C. only scheme. Currently it is 12% above the salary threshold. I believe this is an indication of where it would likely be initially if USS was switched to D.C. only.

Thanks for the link! I don't agree with the logic that employers would not contribute 17% to a DC. It makes no difference to the cost of employment to give 17% (or 16.5% if 2.5% is what goes towards the deficit) to a DC or DB scheme.

Southend1 wrote: »

Your appetite for risk is yours alone, but i can not afford to take that level of risk with my retirement funds. I am relying on USS for the bulk of my income in retirement, other than the state pension. A few years of poor returns pre retirement could mean I have to work many extra years or struggle.

It is true that appetite for risk varies between individuals. But I do not agree that a 30 year investment (largely in the stockmarket) carries significant risk - historically it has been shown that quite the opposite is true and in the long term equities are the least risky investment. The situation may be different for someone much closer to retirement age.

Also, there is different types of risk in a DB scheme. The risk, for example, that you drop dead the day after retirement or well before expected. In the case were there is no dependent or spouse the pot simply disappears, or if there is a spouse they will have a dramatically reduced income.

Southend1 wrote: »

Pooled investments also benefit hugely from economies of scale - individuals in D.C. funds generally switch to low risk investments in the ten years or so before retirement- this protects from investment risk to an extent but also wipes out potential for returns that could be earned in a pooled D.C. or DB scheme.

This is misleading as it gives only part of the picture. The flip side of what you are saying is that the pension pot is invested for the earlier periods in lower return investments. Since this is the bulk of the time, this could be interpreted as a negative of being in the DB.

Southend1 wrote: »

I would not be happy for members to have a choice of a D.C. only option. Not only would it mean many of those who choose that option could well be worse off in retirement, it would undermine the employer covenant for the DB section and hasten its closure, making all of us potentially worse off in retirement.

This sounds rather selfish. I am perfectly capable of coming to my own informed decisions. It seems you would would rather members were stuck in what they feel are inferior pension products so that there is less risk for you personally.

hyubh

Sam_J wrote: »

So it seems that a decent chunk of the 8% + 18% that is going into "my" pension is in reality going towards paying the pension of members in the more generous final salary scheme, to which I have no access.

I'd imagine the scheme documents what proportion of the headline contribution rate is for 'past service' (i.e. deficit contributions) and what is for 'future service' (i.e. contributions for pensions being earned currently). However, referring to the future service rate only still wouldn't really give you a figure comparable to a DC scheme one, for various reasons -

- Headline DB contribution rates are usually not age specific, but the underlying calculations will take account of that, i.e. younger members (even with a CARE scheme) will usually be considered less costly by the actuary compared to older members. As such, use of a common contribution rate involves a slight subsidy of older members by younger ones.

- The USS has a strongly 'grouped' approach to employer costs, i.e. employer rates are mostly determined in common. This leads to institutions that could plausibly be seen to have a much better 'covenant' (i.e. much better placed to stump up extra cash in decades time) effectively subsidise employers with a plausibly weaker covenant.

- Probably stating the obvious, but the key cost of DB over DC is the DB promise. An employer rate of (for sake of argument) 15% could easily become 30% retrospectively were the employer in the future needing to plug an unexpected deficit in the pension fund.

However, I think for members who have only just joined the scheme that it would actually be beneficial to support a move to a fully DC scheme, provided employer contributions remain similar.

Possibly, however...

That way we are not funding the final salary pensions of more senior academics

... not for that reason, given the final salary section is completely closed already. Closing the current CARE section to future accrual wouldn't make the deficit suddenly disappear; the fact contributions are all rolled up into a single rate currently is purely for administrative ease. E.g., the USS could in principle split off past service contributions into distinct capital sums.

hyubh

Sam_J wrote: »

Ideally what I would like is for the union to ask the USS to give members of a choice between staying in the DB or moving to a DC with similar employer contribution.

Keep in mind that if this happened, the DC contribution rates might become tiered by age. E.g., the civil service has operated a choice of very good quality DB or very good quality DC for many years, and the employer rates of the DC section ('Partnership') are tiered by age in order to be actuarially neutral with the CARE section ('Alpha') benefits.

Sam_J

hyubh wrote: »

Keep in mind that if this happened, the DC contribution rates might become tiered by age. E.g., the civil service has operated a choice of very good quality DB or very good quality DC for many years, and the employer rates of the DC section ('Partnership') are tiered by age in order to be actuarially neutral with the CARE section ('Alpha') benefits.

That's very interesting, thank you. I'll check it out!

Sam_J

hyubh wrote: »

I'd imagine the scheme documents what proportion of the headline contribution rate is for 'past service' (i.e. deficit contributions) and what is for 'future service' (i.e. contributions for pensions being earned currently).

Yes, turns out that is 2.5%.

hyubh wrote: »

However, referring to the future service rate only still wouldn't really give you a figure comparable to a DC scheme one, for various reasons...

More interesting stuff. I shall have to digest this. I may come back with a few questions later.

hyubh wrote: »

- Probably stating the obvious, but the key cost of DB over DC is the DB promise.

This is actually something that bugs me. It appears not to be a promise at all. For example, the old final salary members appear to have been shafted. There were making contributions on the assumption they were accruing a proportion of their salary at retirement. Instead, when the scheme was closed last year, their final salary was deemed to be their salary on the day the scheme closed, effectively reducing in value the previous contributions by being forced into "retirement" (for the purposes of the scheme) when they had not retired. DB promises are promises that can be broken.