savings and retirement, and the absence of a guarantee. We save
throughout our working lives, but we require retirement income to
begin many years into the future, e.g., a twenty-five-year-old would
need retirement income some forty to forty-five years later. The
individual investor has just a single lifetime, with an uncertain life
expectancy, over which to execute these decisions and manage this
risk between accumulation and desired retirement income.
In this paper, we focus on the individual investor because that is
the most interesting problem. The extension to institutional investors managing DB plans is trivial.
1 The interesting point that
Modigliani and Muralidhar (2004) emphasized is that if this experiment of transferring risk and these complex decisions to individuals fails, then government bears the cost because it will be compelled to bail out poor retirees. This has interesting implications for
the solution we propose below.
We argue that a key instrument is missing in financial markets—we
call them Bonds for Financial Security (BFFS). Unless governments
(at a minimum) create and issue these bonds, sooner rather than
later, retirement planning will remain a tough challenge for most
citizens. The BFFS is simple to create and we demonstrate how it
achieves the goals of government and average investors thereby
( 1) simplifying investment decisions and eliminating error-prone
approaches; ( 2) reducing the risk of achieving a target retirement
income; ( 3) lowering the costs of investing assets; ( 4) ensuring that
both accumulation and decumulation can take place under the same
entity; ( 5) enjoying complete liquidity and lowering the risks to participants of leaving money on the table if they buy annuities and die
early; ( 6) lowering the risks (and future costs) to the government;
and ( 7) mitigating the current savings glut problem by converting
some pension savings into government infrastructure investments.
Current Assets and Approaches
We assume that all decisions are made in real terms and ignore any
tax implications to simplify the exposition. We also assume that
participants seek a targeted, inflation-indexed, guaranteed income
stream from the date of retirement through death. To achieve this
goal, an individual needs to decide: ( 1) how much to save, ( 2) how
to invest (which incorporates three subdecisions: what assets to
invest the savings in, how much to invest in each, and how to
rebalance), and ( 3) how best to decumulate the savings. Bodie et al.
(2008) try to show how these decisions can be made in an integrated manner, but this is a complicated set of decisions for an
individual to make in dynamic markets with a lot of uncertainty.
After reviewing the shortcomings of current offerings and theories,
we conclude with a brief review of an approach called collective
DC that tries to overcome the challenges posed. It also addresses a
key result first noted in Modigliani and Muralidhar (2004): A guaranteed return on contributions ensures a DB-like outcome.
Why Save or Hold Assets? The Life-Cycle Hypothesis
The fundamental question is: Why does an individual or an institution save and hold a portfolio of assets? Modigliani and Ando (1963)
demonstrate through a simple model that given the life-cycle
hypothesis (LCH), individuals accumulate resources during working life to be able to finance consumption in their non-working life
or retirement. The goal of the individual in LCH is to create a
retirement income stream to finance retirement consumption. In
practical parlance, the desired retirement income stream is referred
to as the “liability” or “L.” The fact that some part of current
income is saved for future consumption/income creates an asset
pool on which investment decisions must be made to ensure that
they are adequate for the retirement phase.
Figure 1 illustrates a simple LCH model with no uncertainty. The
investor is assumed to earn $100/year for forty years, and then
retires in year 2055, with retirement extending twenty years, and
she dies in 2075. She consumes $85/year for sixty years until death.
The black arrow highlights when retirement takes place. The blue
line rising up from zero and growing at a particular rate of return
is the accumulation (or assets) measured on the right-side axis.
Looking at the liability in isolation, it exists even before the individual starts to earn income and is essentially nothing more than
the line highlighted in gold in figure 1—zero until 2055, and then
$85 every year from 2055–2075 (or until death). To generate this
retirement income stream for this level of lifetime income (i.e., an
85-percent replacement rate on final salary) and savings (i.e.,
15 percent of salary), assuming no bequeathing and no discounting
for simplicity, the portfolio must earn a 3.2-percent return per
annum. Note that 3.2-percent annualized does not appear to be a
challenge, but at present most government bonds in the developed
world do not deliver this yield for current maturities, let alone over
an individual’s working life. Risk-taking is therefore a must, and
this exposes the individual to the risk of not achieving the target
The peak-shaped solid blue line measured on the right axis captures
the accumulation of the annual savings that is growing at this rate.
This rate of return was established by asking the simple question:
Given a desired post-retirement income stream (often referred to as
Figure 1: LCH–Accumulating Assets to Finance Retirement
2015 2020 2025 2030 2035 2045 2040 2055 2050 2060 2065 2075 2070
Simple LCH Model
Post Retirement Income