Traditional glidepaths focus on lowering investing risks at preset ages throughout an individual’s life. This algorithmic asset switching targets a particular level of wealth at the retirement date.

However, the relevant risk for retirees is income uncertainty during retirement. The authors posit that focusing on achieving sufficient income through retirement with the use of personalised glidepaths is a better approach to retirement planning than is the use of traditional portfolio glidepaths, such as target date funds.

The authors model the likely retirement income outcomes of  hypothetical individuals through their working life and a period of retirement during which they draw income from their portfolios. The relevant variables used in the modeling are:

• Duration of the working life
• Length of retirement
• Starting salary and annual growth of pay
• Percentage of salary contributed to a retirement plan
• Annual returns and standard deviations of portfolios of stocks and bonds
• Correlations of the asset classes with each other
• Income required in retirement
• Tax rates

In the base case analysis, the authors assume the person earns $40,000 a year at the beginning of her working life and that pay would grow by an inflation-adjusted 1% each year thereafter until retirement at age 65. The individual would contribute 10% of her salary to a retirement portfolio each year. Taxes are assumed to total 15% of the contributions plus the nominal earnings from the investments. The authors also assume the individual would withdraw a certain amount of money from her investment fund each year during retirement based on her individual requirements.

The authors choose the S&P 500 Index and the Barclays US Investment Grade Bond Index to represent stocks and bonds, respectively. These are the only two available investment alternatives in the model.

The authors run Markov chain Monte Carlo simulations that measure the likelihood that there would be sufficient income through retirement under a full range of glidepath alternatives. From this analysis, the authors estimate the probability that an individual would be in financial peril. The authors also conduct sensitivity analysis in which they alter the base case variables. For instance, they examine what would happen with different assumptions for asset returns, correlation, and volatility.

In addition to sensitivity analysis, the authors draw attention to the fact that the model inputs can be personalised to reflect different career paths. “Few individuals experience a continuous working life followed by a smooth transition to retirement,” they state. A key advantage of this approach is the ability “to map an individual’s glidepath as circumstances change, so that suboptimal asset allocation can be avoided.”

The authors use the example of a person who sees his income jump by 100% in his mid-thirties and then use the modeling results to estimate how likely it would be for him to be in financial peril during retirement. Likewise, the authors indicate that their approach allows for the modeling of someone who leaves the workforce for a few years to rear children.

Staying heavily invested in stocks as long as possible before retirement significantly reduces the risks that a person will run out of money during retirement. For instance, the authors provide an extreme example that indicates that an immediate switch from 80% stocks/20% bonds to 20% stocks/80% bonds before retirement makes the risks of financial peril high: “The baseline assumption [80/20 stock/bond allocation] means that the average worker could accumulate sufficient wealth to avoid financial peril 80% of the time.”

The authors demonstrate that they could model income structures that change radically during an individual’s career. They use the example of someone whose pay doubles at age 35 and find that the 80/20 stock/bond mix works best, although a split above 50/50 still leaves only a modest risk of peril.

Tailored glidepaths for individuals are possible and reduce the risk that individuals run out of money during retirement. “If we can feasibly customise glidepaths for individuals, we should,” the authors state. “Construction of glidepaths for individuals offers workers optimised portfolio outcomes focused on retirement income sufficiency.”