Joe is in a quandary. He had figured that he would buy a new truck in 3 years. He hadn’t been paying attention to the news much and assumed his $40,000 savings would grow by 3% annually so he estimated he would have approximately $44,000 to spend. In fact, given current interest rates, his interest rate on his savings is close to zero.
Joe could respond by saving more money and reducing his current spending elsewhere to ensure he still has $44,000 to spend. Alternatively, he could note that there is no incentive for him to delay his purchase and he might as well enjoy his new truck right now.
Faced with low, possibly negative interest rates1, households planning for retirement face a similar dilemma. With a combined saving and retirement period more likely to be 3 decades, rather than 3 years, and retirement assets of several hundred thousand dollars, considerably more is at risk for many households compared to Joe.
A recent U.S. study suggests that the advice frequently offered by financial advisers to investors facing low interest rates is at odds with the same model advisors use for explaining investor behaviour. To see how this arises we consider the impact of interest rates on the time value of money and introduce the life cycle model of consumption.
As noted above, low interest rates can cause households to reduce their savings because of lower rewards, a behaviour known as the substitution effect, or to increase their savings to maintain their retirement income, known as the income or wealth effect. Which behaviour dominates depends on comparing the interest rate to the subjective discount rate. The subjective discount rate is a measure of impatience: all other things being equal, we expect to be paid for waiting. For example, Joe might have a subjective discount rate of 3% and his behaviour is dominated by the substitution effect when he realizes interest rates are less than 3%.
The life cycle model of consumption states that investors aim to smooth consumption over their lifetime in the face of varying income. For retirees this means saving enough to ensure that consumption (typically your lifestyle expenditure, excluding savings and assuming the mortgage is paid off) doesn’t change when you retire.
The study focuses on late career workers who have 10 years of employment before retirement. Subject to a fall in interest rates, how do they respond if they are following the life cycle model? The first observation is that reducing interest rates reduces their total expected wealth. With less total wealth, consumption should fall in every period, whether working or retired.
The second observation is that the price of later consumption has gone up relative to the price of current consumption. To illustrate, suppose I want to spend $10,000 in 10 years. With an interest rate of 3% I would have to set aside $7,444 today. At an interest rate of zero I would have to set aside $10,000, an increase of 34%. In other words, the cost of being able to spend $10,000 in 10 years has increased.
The study showed that, after an initial modest increase, consumption decreased. When interest rates fell by 2%, the average decrease in annual consumption was 16% over the remaining life of the investor.
A fall in interest rates impacts savings decisions by lowering total wealth and increasing the preference for consumption now rather than later. In the study, the net impact is to reduce overall consumption and to reduce consumption near the end of retirement relative to the early years of retirement. This is at variance with financial advice that recommends greater savings to maintain consumption constant at the same level prior to the decline in interest rates.
Of course, the lifecycle model is just a model, not a prescription, although there is considerable evidence that households do behave according to the life cycle model. In the face of declining interest rates a late career worker or retiree may decide that, despite the wealth they want to save more to maintain their level of retirement consumption at the same level prior to the decline in interest rates. For the late career worker with only a few years to retirement this may mean saving a lot more in the few remaining years of retirement. Since increased savings must come at the expense of reduced consumption when working, the implication is a jump in consumption at retirement, contrary to the goal of the lifecycle model.
Independent of the level of interest rates, there is evidence across the developed economies that spending declines naturally with age (at about 1% annually, after age 70) due to declining health and mobility.
Other options to boost retirement income as interest rates decline are:
A European study looked at savings behaviour in response to changing interest rates. Overall, the study found that savers respond to lower interest rates by increasing savings, but the relationship is weak: for every 1% decline in interest rates savings rates increased by only 0.2%. The study points out that for many citizens of European countries short term deposits are a large proportion of total savings and should be expected to be especially sensitive to interest rate policies.
Other factors such as changes in health expenditure had a much larger impact: for example, a 1% fall in health expenditure lead to a 1.5% increase in savings rate.
Japan, which has endured decades of low interest rates and a high allocation of savings to short term deposits, leads the developed world in the unenviable ranking of retirement savings deficit, even before Covid-19. The World Economic Forum (WEF) reported that Japanese retirees will outlive their savings by 15.1 years for males and 19.9 years for females. The comparable data for Canada is 9.9 years for males and 12.7 years for females.
A low interest rate world exacerbates the challenge for retirees but needs to be addressed by understanding the total financial circumstances of the client and the overall economic circumstances. We have seen that rather than simply advocating a higher level of saving, a more nuanced approach recognising both the decline in real wealth and an increased preference for spending sooner rather than later, is more consistent with investor behaviour.
Sound advice, consistent with our understanding of how investors actually behave, should surely be part of the solution. Indeed, a UK study concluded that “those who take advice are likely to accumulate more financial and pension wealth, supported by increased saving and investing in equity assets, while those in retirement are likely to have more income, particularly at older ages”.
In conclusion, we echo the authors of the WEF study who stated, “Aligning the interests of the adviser and the saver is perhaps the most significant challenge for all countries. Conflicts of interest can be a significant impediment to achieving good retirement, from either the direct effect of “bad”, inappropriate or costly advice to potentially reducing trust in the advisory model, thus leading to fewer people taking advice despite needing it.”
