Defined Benefit vs. Defined Contribution: Analysis

I've run a few different scenarios to work out whether I should go for the defined benefit or defined contribution option.

Scenario 1: Work 1 Year at an Australian University

This is the simplest scenario as the only variable is future rates of return in the defined contribution scheme. I don't need to project my future salary. I assume that the tax on superannuation earnings averages 7.5% (15% on general income, 10% on capital gains, and zero on fully franked dividends). All numbers are in 2009 Australian Dollars. I don't worry about projecting inflation. I am currently 44 years old.

The results of the analysis are also very clear cut. I and my employer would contribute a total after tax amount of $13,262 this year.

Under the defined benefit scheme I would receive $13,603 if I retired at age 60 (the earliest age I could access the benefits) and $14,221 if I retired at 65 or older. Clearly this rate of return is very low. A 1% real rate of return gives a lump sum of $15,450 at age 60 and $16,181 at age 65 in the defined contribution scheme. A 5% real rate of return gives $28,460 at age 60.

Selecting this option only makes sense in this scenario if you are extremely averse to market risk. Unless you expect hyperinflation and negative real interest rates in the future it will make more sense to invest in the cash option in the defined contribution scheme.

Scenario 2: Work Till Age 60 at Current Salary

Under this scenario total contributions are $225k and the age 60 defined benefit is $231k. Again, a 1% real rate of return beats the defined benefit.

Scenario 3: Smooth Rise in Salary to Full Professor (E1) at age 60

An E1 Professor currently earns $133,901. This scenario is not clear cut. The defined benefit is $505k if taken at age 65 but actually quitting at 60. A 3% real rate of return gives $508k. Continuing to work at that salary till age 65 favors the defined benefit a little bit more.

Scenario 4: Switch to Full Time Next Year and Then Smooth Rise to E1

Now the age 65 defined benefit (working till 60) is $513k and a 2% rate of return gives $516k.

Scenario 5: Switch to Full Time Next Year and Fast Promotion to E1

I assume I rise one salary notch every two years. I become a full professor in ten years. Defined Benefit is $523k at age 65 and the 2% rate of return yields $535k.

As you can see, the earlier promotion comes or if promotion doesn't come at all the lower the require rate of return in the defined contribution scheme. I reckon scenarios 1 and 5 are most likely. I'd assign them a 40% probability each and the other 3 scenarios (20/3)% each. The expected value of the defined benefit is then $299k of a 1% real rate of return defined contribution it is $273k, at a 2% real rate of return $309k , at a 3% rate of return $350k.

Using expected value assumes I am risk neutral. If I am averse to career risk then I should put a heavier weight on Scenarios 1 and 2 than on the more positive career scenarios. Those scenarios have lower required investment rates of return.

Given this low required rate of return and the advantage of portability I am going to choose the defined contribution scheme.

Defined Benefit vs. Defined Contribution

The major decision I need to make in response to receiving my job contract is which type of superannuation (retirement) scheme to join. Yes, we have a choice between defined benefit and defined contribution (or accumulation in the Australian jargon). Defined benefit is mainly based on your average salary in the last three years that you work for a university that is a member of the Unisuper fund (indexed for inflation) and the number of years that you contribute to the fund. So here the main risk is career risk. This option performs best for someone who will work their whole career at Australian Universities and get promoted to professor or dean right at the end of their career. In defined contribution you have both a career and a market risk but much more of a market risk. An important benefit of the defined contribution is that you can rollover your benefit into another superannuation fund - i.e. it is portable - while the defined benefit is not.

When I previously worked at an Australian university we initially had no choice and were all in a defined benefit scheme. Then when we were given the choice of switching to defined contribution I did it immediately as I didn't expect to stay in the Australian university system in the long-term. After I left the university I rolled my super into Colonial First State. That was all going well until some unfortunate investment decisions last year.

Given my great career uncertainty at this point it seems obvious to go for the defined contribution scheme. But I will do a proper analysis and report back with the results.

The Unisuper scheme has, on top of a massive 17% employer contribution, a required 7% contribution from the employee's nominal salary. To balance things out, and given our higher income now, I will start 7% "salary sacrificing" (pre-tax employee contribution) for Snork Maiden too. Her employer contributes 15.4%. As her salary is higher, it seems fair to me to go for the equal percentage salary sacrifices.

BTW I'll see no gain from salary sacrificing this tax year as my marginal tax rate is 15% which is the same as the superannuation contributions tax.*

* In Australia retirement contributions are usually taxed at 15% going into the fund. You can choose to make employee contributions pre- or post-tax. Post-tax ones don't attract the contributions tax but obviously you pay your regular income tax on them. If your marginal tax rate is above 15% (as most people's is) then it seems like a no-brainer to go for the pre-tax contribution known as "salary sacrifice".

Was It a Good Decision to Switch to Defined Contribution Superannuation?

Back in 2009 when I started with my current employer, I carried out a cost-benefit analysis to see whether it made sense to stay in the default defined benefit scheme or to switch to the defined contribution scheme. As a result of the analysis I switched to defined contribution.

Was that a good decision. Using the info in the Unisuper PDS and my data I compute that if I retired at the end of this month I would get a lump sum of AUD 213k. My actual Unisuper account is at AUD 284k. So, so far it's been a good decision.

For context, in Britain, there have been strikes and demonstrations against the plan to switch academics from  defined benefit to defined contribution. But I see defined benefit as a regressive form of socialism where people who are promoted near the end of their career suck the benefits from the system. This is because the lump sum benefit is proportional to the members salary in the last 5 years. I've seem quite a few people promoted to professor in their last few years and of course, deans and other senior administrators benefit heavily from the scheme. This is at the expense of successful researchers who are promoted early and stay in research at a more or less constant salary.

Defined Benefit vs. Defined Contribution Update

Each time I was given the choice to be in a defined benefit scheme or a defined contribution superannuation scheme, I chose defined contribution. So, now and then I like to check whether I made the right decision. I worked from 1996 to 2001 in the Australian Higher Education sector. In fact, at the same employer I now work for. Originally, defined benefit was the only option. But then they gave us the choice to switch. In 2001, I rolled over my account to Colonial First State and then this year to our SMSF. I have now worked out how much that money is now worth. I estimate that it is AUD 410k. In 2009 I started working at the same employer again and opened a new Unisuper defined contribution account. It now has AUD 477k in it. So, in total I have AUD 888k. Including the 5 years that I worked from 1996 to 2001 my defined benefit lump sum would now be AUD 473k. We are going to need to have a big crash to make those numbers equal...

Thanksgiving Conversation

As many other bloggers mention, we often find ourselves having conversations about finances with people we don't usually talk to about finances at occasions like Thanksgiving. And it is always amazing what people don't know. We were talking about looking for other jobs and relocating and someone mentioned: "It's a pity that retirement accounts are set up in America so that it makes it difficult to move jobs". With a bit of prodding it sounds like the guy has a 403(b) defined contribution account with TIAA-CREF just like I do. I told him that it probably isn't a barrier to moving but he should check with his HR people if he is thinking of moving so he knows what to do. I said anyway he could roll it over into an IRA even if he couldn't keep the account or transfer it to his new employer and then he could just start a new account with his new employer. Our host (early 50s, actually with a PhD in economics) then said: "how do you know if you have a defined benefit or defined contribution?" I explained, looks like she has defined benefit with state government and that could be an impediment to moving. If you stay with the state more than 20 years they up the benefit level and she's only been 16 years. That's what she said anyway. Then everyone commented that they hate thinking about this stuff and dealing with money... I didn't investigate why.

The first guy though did know how much was in his 403(b), how much he contributed each month, and had shifted some of the money to more aggressive options in the previous year. The host had an FSA (something I don't do because I think it's too much hassle) and knew all these things came out pre-tax. We also discussed the percentages my employer and I put into my 403(b).

Phoned Unisuper

So I phoned Unisuper and told them I'd sent the form in twice to switch to defined contribution (or "accumulation" as they're calling it) and I'm wondering why I'm still in the defined benefit scheme. The guy I got on the phone told me that everyone starts in Defined Benefit (so they just ignore the initial membership form despite it asking which plan you want to be in) and that they got my switch form in May but "it takes a long time for the switch to happen" and he doesn't know why that is.... He said it was "fair enough" that I was wondering what was happening. So I suggested I'll check in another month whether the switch has happened. Which he said "makes sense".

Today's Moves

I bought more corporate bonds. This time in Royal Bank of Canada and Welltower. These were the highest yielding investment grade bonds that I could actually buy that mature in April. $US25k in one and $US35k in the other. Not ideal, as the commission is higher for the first $10k but those were the amounts on offer. I was looking to buy bonds of Siam Bank Corporation, but the minimum purchase turned out to be $US200k for some reason. I also tried to buy Glencore bonds, but when I put my bid in the market, the offer disappeared from the screen. I waited a while and I didn't get the bonds. Very weird. Probably it makes sense to wait until I have enough cash to buy $US100k and buy Treasury bills unless there is a sufficient amount of corporate bond with a high enough yield to make it worthwhile. After commissions these two purchases probably end up breakeven with a Treasury bill. As I begin to buy bonds at longer maturities though, the commission will be spread out over a longer period. I use the bond scanner provided by Interactive Brokers to find available bonds with the right characteristics.

I also am looking at shifting our allocation in the PSS(AP) superannuation fund from 50/50 "balanced" and "aggressive" to 100% balanced as part of our general de-risking. I am again reminded of how shocking the lack of transparency about investments is for Australian funds compared to US funds. All the information they provide in the annual report is the percentage of the fund allocated to "equities", "alternatives" etc. with no further details. PSS(AP) actually used to provide more, but not a lot more, information than this. An interesting fact from the annual report is that employer contributions totaled $A1.154 billion and employee contributions $A55 million in the 2017-18 financial year. Not many people are making additional contributions or they are not making very large contributions. This makes sense as the employer (the Public Service) contributes 15.4% on top of the official salary to the fund. It's only interesting because for the defined benefit fund at Australian universities – not part of the public service, though they are in the public sector – employees are required to contribute 7% on top of the employer 17% in order to get full benefits. I opted out of the defined benefit fund. Our employee contributions at PSS(AP) are actually as big as the employer contribution at the moment.

P.S.
Basec on reading the Unisuper report, employee contributions might only include non-concessional or "after tax" contributions and not salary-sacrificed or "pre-tax" contributions. This is because the stated contributions tax in the report is 15% of the employer contributions. By contrast with PSSAP though, Unisuper defined contribution members make massive non-concessional contributions (see p52 of the report), even though the employer makes 17% contributions to the fund.

Superannuation Handbook 2007-08

I just finished reading The Superannuation Handbook 2007-8 by Koken and Smith. It is a pretty comprehensive coverage of Australia's very complex superannuation or retirement system. I think my understanding of the system improved somewhat after completing the book, though it is still hard to keep all the facts and rules organized in my head.

Australia's superannuation system is complex for a number of reasons. First and foremost, governments have continually changed the rules while trying to grandfather in existing super investors in many cases. And there have been very significant recent changes. In the US, new rules have often meant the creation of new types of retirement account such as the Roth IRA or Roth 401k. In Australia there is only one type of account and all the various rules have been applied to that same account class. So we have pre-tax and post-tax money going into the same accounts, for example. In the US defined benefit pension schemes are an entirely separate beast to defined contribution retirement accounts. Not so in Australia.

Second, while the US does not tax money in retirement accounts Australia does (the US taxes payouts from accounts that had pre-tax contributions like the 401k). This I suppose is why self-managed superannuation accounts in Australia are subject to such a bureaucratic regulatory nightmare compared to IRA accounts in the US.

Third, there are several different age thresholds (55, 60, and 65) at which investors have different rights to access their super and varying taxation obligations if they do. In the US there is a single age threshold of 59 1/2, though you can access your money before then subject to tax and penalties (unless you do a 72t or annuity).

Fourth, eligibility for social security in the US does not depend on assets whether in retirement accounts or not. Access to the age pension and other benefits in Australia does depend on income and assets tests and sometimes it matters if the source is from super or not (but less than in the past).

Fifth, in Australia, how much tax you pay depends on how you take the money out of your super account - whether as a variety of different "income stream" products or as lump sums.

Well, there are probably more reasons that don't immediately come to my confused mind that result in the Australian system seeming more complex to me.

The book does an admirable good job of covering this very confusing topic. There are three points though which are somewhat weak. Not all terms are clearly defined. For example, the entry in the glossary just says that a "complying pension" complies with certain regulations. It'd be nice to spell out some of these more clearly.

Second, the authors often gloss over details and technicalities. Footnotes or appendices to chapters could cover these if they don't want to complicate the text further. For example, there is a rule that a low-income self-employed person cannot get a government co-contribution if their "business income" is less than 10% of their total income. In other words the rest of their income is from investments or superannuation etc. This is the kind of point that was glossed over that I wanted to get a straight picture on.

Third, there are plenty of worked examples in the text, but most of these only cover the first year of any investment program. In some cases they comment that the difference between investing in superannuation or outside superannuation isn't that big. But that's the result after only one year. The results of investing in superannuation or outside superannuation could look quite different in the long-term than in the short-term.

Bottom line, I'd recommend this book as a very solid background to the topic though you might need to consult the ATO website and other resources along the way.

Future Fund Allocation

The Australian Government has a sovereign wealth fund known as the Future Fund. The Commonwealth Government has run budget surpluses for many years. Once they had paid down most of the national debt they started to accumulate surpluses which in 2006 were allocated to this fund and dedicated to funding previously unfunded defined benefit retirement payments to public servants. The Future Fund's website gives some information on their investment policies:

Equities: 35%
Tangible Assets: 30%
Debt: 20%
Alternative Assets: 15%
Cash: 0%

Based on the PSS(AP) superannuation fund we can guess that 7% is allocated to US equities and therefore 28% to non-US equities. We can then compare this portfolio to the others I already discussed.

Both the PSS(AP) portfolio and Moominmama are almost as close to this allocation as they are to the average US university endowment fund. Moom is slightly closer to this portfolio than to the average US endowment but quite far from both (39% and 43%). Compared to PSS(AP) - which of course is an Australian government defined contribution retirement fund - the Future fund has double the allocation to debt and to real assets and is much lighter in equities and cash.

Second Thoughts on Retirement Accounts?

As I begin to realize that maybe I can realize my dream of financial independence much faster than I ever thought possible, I am beginning to rethink the role of retirement accounts in my financial plans. Up till recently I have always contributed the minimum to retirement plans with my employers. The other side of the coin is that my main employers have had extreme levels of retirement contributions. In my first academic job in the UK contributing to the retirement plan was voluntary. It was defined benefit and I was only planning on staying a year or two. I thought reducing my debt was more important. The cost-benefit analysis I did at the time showed that the right choice was to opt out. In my next job (in the US) I didn't stay long enough to become eligible for the retirement plan. Then when I moved to Australia, participation was mandatory. Between employer and compulsory employee contributions, 24% of my nominal salary was going into my retirement account. I thought this was enough especially as it is almost impossible to get the money out in the Australian system before age 60. Then after a gap where I claimed to be "self-employed" :) I was back in the US and here we have to contribute a minimum of 1% and the employer 8%. My goal was to rebuild my savings from the 2002 drawdown and then head for the goal of financial independence. Then this year I opened a Roth IRA account when I learned that you can always withdraw the contributions without penalty and use up to $10,000 in earnings towards the first time purchase of a house. My plan was to save towards a potential downpayment using a Roth IRA. This summer, as I began to understand the tremendous profit potential of my trading model I began to rethink my Roth plan. I think I will hardly miss the $4000 annual contributions and the earnings on them. Therefore, I now plan to continue contributing as long as I can and not withdraw my contributions as soon as I hit $10,000 in profit (I am halfway there at the moment).

Now a new idea has struck me. If I maximized my 403(b) contributions to the $15,000 annual limit I could rollover my account into my Roth when I eventually leave my current employer (subject to paying the appropriate taxes). This would be a way of getting much more money into the tax-free Roth environment. I think I may do this as soon as my trading program has reached its full size and if I continue to be profitable money will then be flowing out of the trading program into long term investments (in order to minimize the potential of a catastrophic loss there is a maximum sensible leverage that should be used in a trading program relative to total net worth - in the long term probably 70% of net worth will be directed to long-term investment and 30% to margin for trading). At that stage I will no longer need my salary to expand the size of the trading program and can direct it to a more tax advantaged environment.

The bottom line is that if you are following a Kiyosakian path at first you want to maximize the assets you have to achieve immediate financial independence. But once that goal is achieved, it makes sense to take full advantage of tax sheltered retirement accounts.

Unisuper Member Pack Arrived

My membership certificate and other info finally arrived from my new superannuation fund, Unisuper. They completely ignored all the choices I made on my application form (apart from noting my tax file number - equivalent of the SSN in the US) and so now I need to submit them again. Including switching to the accumulation plan from the defined benefit plan. This seems to be par for the course for every retirement plan I or Snork Maiden have joined. No wonder people tend to stick in the default options provided by retirement plans!

I also added $A2,000 today to Snork Maiden's non-retirement account with Colonial First State and paid off another $A2,000 of my margin loan with CommSec. It's nice to have money to save again!

Income of American Retirees

Interesting discussion in the comments on a Krugman post on retirement income based around this pie chart of income sources of Americans of 65 and older in the 75% to 50% quartile (second from top):



The chart shows that only 9% of this group's income comes from assets - i.e. 401k's, taxable accounts, rental housing etc. But some major caveats are needed in order to understand the results:

1. The asset income does not include capital gains or drawdowns of principal. In reality this group is far more reliant than this on assets they own for their income.

2. The data is for 2008 when interest rates were hitting record lows and dividends were being cut. Income in previous years would have been higher.

3. Defined benefit pensions rely on underlying investments in capital assets, usually the retirees are not exposed to the fluctuations in the underlying investments unless the plan ends up collapsing due to underfunding...

By the way, here is the income sources for the top quartile:



They do get a greater share of their income from assets but they are also working more or more of them are working.

All of Labor's Tax Increases

The Labor party is at the moment likely to win the next federal election in Australia in May. Labor has become increasingly left wing in recent years and has a long list of policies to raise taxes. This is, I think, a comprehensive list:

1. Abolish Liberal plan to raise the top tax threshold to $200k: This was supposed to happen in 2024. The top tax bracket will still cut in at $180k (about USD130k) where it has been for many years. Bracket creep is pushing more and more taxpayers into the top bracket. This will affect us if I am still working then. If I'm not, probably my taxable income will be lower.
   
2. Raise the top tax rate: Add 2% to the top rate to raise it to 47%. With Medicare that is 49%. This will immediately raise our taxes.
   
3. Abolish plan to eliminate 37% tax bracket: This also was supposed to happen in 2024, so may not affect us except to the extent of how many franking credits will get used up offsetting our taxes, if I retire by then.
   
4. Repeal already-legislated tax cuts for companies with turnovers of between $10 million and $50 million: Small businesses pay 27.5% corporation tax and larger companies 30%.  The government wanted to extend the low rate to larger companies. This is unlikely to directly affect us.
5. Reduce the long-term capital gains tax discount to 25%: The discount is now 50%. This will have an immediate impact on us as we have run out of accumulated tax losses. OTOH existing investments will be grandfathered. It makes it more attractive to incorporate and pay CGT of 27.5% instead of 37.5%.
   
6. Abolish refundability of franking credits: Since 2000, if you have excess tax credits from Australian companies beyond those that offset the taxes you need to pay you can get a cash refund. I did benefit from this once or twice soon after we moved to Australia and my income was low. This will have a big impact on superannuation funds in pension phase that have zero tax to pay and possibly even in accumulation phase if they have a lot of franked dividends. It will affect lower income self-funded retirees with money outside superannuation too.  Some listed investment companies (closed end funds) are already paying out special dividends to get franking credits out of the fund and to investors before the end of the financial year. On the other hand, I don't think these funds will radically restructure due to this proposal. I don't think it will have a big impact on us as I've planned to put the least tax advantaged investments like managed futures into our planned SMSF. And I expect we would be in the 32.5% tax bracket when retired. If I retire at 60 say and start a superannuation pension we could use franking credits inside our SMSF to offset Moominmama's superannuation earnings tax liability as she is 10 years younger. And then maybe we could add Moomin to the superannuation fund :)
7. Abolish negative gearing: This is the ability to deduct investment costs beyond the earnings of an investment from other income. This mainly applies to property investors who mostly lose money in Australia in the short run, hoping for a long-run capital gain. We don't negative gear so it shouldn't affect us. Wealthier property investors who also own shares or other investments will be able to offset their losses in property against dividend and other income. So, like many of the Labor measures they mainly hit lower income investors...
   
8. Tax discretionary trusts as companies: These are trusts that have multiple beneficiaries and can alter what earnings they stream to which beneficiary on a year by year basis. Actually, they are proposing to tax trust distributions at a minimum of 30%. So, it's not like a company which pays 27.5% tax in the case of a small business and then distributes franking credits. I don't see any justification for allowing this kind of tax dodging. However, I think they should just require all trusts to be unit trusts with defined shares and everyone sharing in all income. These operate just like unlisted managed funds (mutual funds). I think most discretionary trusts will just do this if it's allowed.
   
9. Reduce annual non-concessional superannuation contributions to $75k: This would mean it would take us more years to make all the non-concessional contributions we want to make and means I probably should already get one in this financial year.
   
10. Reduce the threshold for 30% superannuation contributions tax to $200k: Currently the threshold is $250k. The threshold includes employer superannuation contributions, so this will definitely affect me.
    
11. Remove the right, already legislated by the government, of superannuants to make catch-up contributions when their super balance is less than $500,000: I don't think this is probably a big deal. It will mean stretching contributions over more years.
    
12. Reduce ability to take tax deductions for additional concessional superannuation contributions: People will need to have 90% of their income or more from sources other than employment to do this. I don't understand why concessional contributions for employees are limited to salary-sacrificed contributions and you can't make more concessional contributions unless you really aren't an employee. The Liberals tried to fix this anomaly.
13. Limit tax free pensions to $75k per year: Currently you can transfer up to $1.6 million into an account to fund a tax free superannuation pension. At a 4% initial withdrawal rate (required rate for under 65s) that is $64k per year. At 5% (65-74 y.o.) it is $80k per year. So, Labor's proposal is not that restrictive. However, if the $1.6 million earns a lot more than that a year, it will be taxed a lot more than at present.
14. Limit deductions for tax advice to $3,000 per year: I am assuming that this won't apply to companies or superannuation funds, just to individuals. In which case, it isn't a big deal.

I think most people are probably aware of one or two of these but don't have a good idea of the extent of the proposed tax increases. A big question is whether Labor will have sufficient control of the Senate to pass all these measures.