Why you always need a model

It is a dangerous fallacy to think that the same investment advice can be given to different people. This is particularly true if individuals are dissimilar in age, financial situation, objectives, or general preference for risk. The importance of models in creating frameworks to guide us, individuals, investors or policy makers, in the decision making process is recognized in the diverse fields of science. A good example is given by a recent piece published by Paul Krugman, an economics laureate, in the New York Times on the importance of models in guiding policy (link).

When working on your financial planning, such things as investment-horizon or your objectives are crucial. If you are planning to save and keep those savings safe for a house down-payment in two years, than your investment portfolio should certainly be a lot less volatile than someone who is saving for retirement, twenty years down the road. Or if you are a recently employed university graduate, a steady employment income should allow you to better absorb temporary downturns in your financial wealth due to capital market fluctuations, compared to someone already in retirement and dependent solely on the past accumulated wealth. Furthermore, while it can be shown that investments with higher risk and return might statistically dominate less aggressive investment strategies over the long term, if a person is very risk averse, all the financial market fluctuations could cause more detriment to this person’s wellbeing than a higher financial wealth at a later stage in life. Hence, aversion to risk is an important component to consider when deciding on investment strategy.

For all the named reasons and a long list of additional ones, financial planning decisions need a guiding framework. Saalt believes that a good tool to compare and check your financial planning is offered by economic models focused on dealing with household finance, more precisely with life-cycle theory. This field of research has for decades analyzed the behavior of individuals and households in their effort to smooth consumption over their lifetime, and it has developed a series of theoretical models that provide a tractable way to optimize this behavior – to learn more, move down to the detailed sections below.

Model – machine (Optimization)

The core of our modelling framework is based on household finance and the life-cycle theory, and a conviction that individuals can and should optimize their financial planning given an array of personal characteristics and circumstances. Nonetheless, some people do not have standard income dynamics. These people include freelancers, artists, non-working spouses, writers, and many others. For those, and others who might not wish to disclose personal characteristics, we have developed an investment targeting tool that works with simpler mechanics.

Target investing

The user is given a tool to visualize probabilities of reaching a targeted investment amount, given a starting investment level, a time horizon, and a choice between different potential expected returns and standard deviation combinations of a broadly diversified portfolio (market portfolio) and a safe asset. Once these parameters are chosen, a Monte Carlo simulation is run, which generates the potential outcome distribution and the probability of reaching the envisaged target.

Household finance and life-cycle theory

Consumption Smoothing

In the early 1950s Nobel Laureate Franco Modigliani began studying the consumption and investment decisions of individuals and households. He realized that people adjusted their spending and saving behavior to both the available resources to them over their lifetime as well as to their current stage in life. It became clear that there is a pattern of building up a stock of assets in the earlier phases of life, which could then be used in the later stages, particularly after retirement, when there no longer was a regular flow of income. This behavior is generally termed “consumption smoothing” in the economic literature.

Within the field of economics, there has always been the tension between the camp focused on determining and estimating the behavior of individuals assumed to be fully rational in their choices, which forms the field of “positive economics”; and those academics who work on creating models to support and correct imperfect choices (“normative economics”). This platform’s approach is to provide users with a model framework, which can generate optimal asset allocation and savings suggestions based on the individual’s characteristics, so as to allow him/her a comparative benchmark and a tool to assess the sustainability of current saving behavior. To a certain extent it keeps up the tension between the two camps by making use of insights from both sides and presenting them in a way which helps individuals in their financial planning decisions.

Labor income dynamics

Labor income as “safe asset”

A different way to look at labor income is to regard it as a safe asset, something an individual can rely on for the time he or she is fit to work. It is a reliable resource because, even if people do not always “land a dream job,” people are flexible and adapt their expectations to existent market conditions so as to make sure not to stay unemployed for too long. Hence, under normal circumstances, people can expect to earn a steady flow of income for the period of their working age.

When thinking about it, a bond has very similar characteristics, it delivers a steady flow of coupon payments until maturity. The pricing of a bond therefore discounts all future cash flows to its present values in order to obtain a measure of how much it is worth today. The same discounting cash flow approach can be applied to a person’s expected income flow in order to obtain the present value of the asset called “ability” or “human capital”. That asset is obviously not tradable due to the risk of moral hazard – i.e. a person selling the asset and immediately thereafter stopping to work – and the impossibility and immorality of contracts which would force a person to work. Nonetheless, for an individual, one’s “human capital” should be treated as a valuable asset.

When holding on to the image of labor income as a valuable asset, it is clear that this ability is worth the most early on in someone’s career, since at that time the majority of one’s paychecks still lies ahead. At the same time, the accumulated financial wealth is still small, since there has been no time to save and invest. The figure on the right hand side presents a summarizing example of this concept: at an earlier age, the person’s wealth lies in his or her ability to work, and, as time goes by, financial wealth rapidly increases its share in total wealth. Therefore, and due to the fact that labor income has many characteristics of a safe asset, it is advisable for a young person to have a larger portion of financial wealth allocated to assets with a higher risk-return profile. This argument holds not only for the aim of trying to have a broadly balanced lifetime risk-profile, but also because a younger person has more time to recoup any potential temporary shock to financial market performance.

Risk preferences

Finally, incorporating someone’s appetite for risk is certainly among the most important components in financial planning. That is the reason why most advisory tools stress this component – although often neglecting other named components above, which in many instances might even be of higher order of importance.

Identifying a person’s risk preference profile is extremely difficult, and a whole branch of academics from the fields of economics, finance, as well as psychology have produced interesting research on this – a small bibliography can be found below. Interestingly, research has found that people tend to have different risk preferences depending on the environment in which the test takes place. But not only that, the type of payoff in a gambling environment also has an important influence on the estimated result. Saalt tries to keep the type and quantity of questions within a tractable framework, using only three questions but designing them to cover different testing backdrops. The output is then averaged and linearly mapped into the width of relative risk aversion parameters used in Campbell and Viceira (2001).

Existent financial wealth

The capacity to smooth today’s and future consumption will be considerably influenced by the existence and size of accumulated wealth. Concomitantly, built-up capital will influence the current and future size of the saving rate. The concept is rather intuitive: if at some point in time a person has built up a considerable amount of wealth, then the need to save is less pronounced. Within Saalt’s modeling context, consumption is a function of income and wealth, as well as other parameters.

The model also finds target ratios of financial wealth to labor income for investors. That target ratio can increase from about three to over twenty. The increase is driven particularly by the degree of risk aversion of an individual. In practical terms, this means that one should expect someone with an intermediate risk aversion profile and some twenty years to retirement to achieve a target financial wealth that is something around 12-14 times the income level. While these ratios seem rather large, it is important to bear in mind that the model assumes all of an individual’s consumption will have to be financed solely by accumulated wealth – there is no expected other income flow once retired. That is why the platform allows a user to input his or her expected additional future income streams so as to allow a greater degree of tailoring.

Using all ingredients to achieve the right balance between market portfolio and safe asset

All the components discussed above are crucial input variables in any serious attempt to tailor financial planning and asset allocation. The modeling framework we have used draws heavily on the great research work by Professor John Campbell, University of Chicago, and Luis Viceira, Harvard University. Particularly their book “Strategic Asset Allocation: Portfolio Choice for Long-Term Investors”, published in 2002, has been enormously helpful in building the right foundations, and it provides a great read for anyone interested in deepening their knowledge in the field. Other research been processed directly or indirectly in this financial literacy platform. The full list of publications can be found at the bibliography at the bottom of the page.

Contextualizing the issues in a specific example

Finally, in order to bring some of these discussed items to life, it is useful to pick a practical example to illustrate the optimization results. The figures below have been generated by running an optimization procedure for a person aged 25, with a university degree, a yearly salary of $84,000, and a financial wealth of $45,000.

The upper left hand side figure shows the expected income trajectory, in line with the estimated income dynamics presented in the labor income section. The model assumes that retirement sets in at age 65 and there will be no pension income. The picture on the upper right hand side shows what the model generated optimal consumption profile should look like for the different types of risk appetite (z-axis/3-D axis), whereby a larger number implies a greater risk aversion. As one would assume, an individual with a greater affinity for risk, will also consume more and save less. The figure on the lower right also resonates that result by showing that more risk-prone individuals should allocate a greater portion of their financial assets into the portfolio with a higher risk-return profile (i.e. the market portfolio). Finally, the figure on the lower left synthesizes all of the information and decisions taken by that individual by showing the different financial wealth trajectories. Clearly, wealth accumulation will depend on income dynamics for a great part, but also on consumption and investment patterns. One can see that individuals with less affinity for risk will tend to accumulate more wealth throughout their working life and will also make sure to have a sufficiently large buffer for any emergency.