The contribution of the elderly to national savings is an important indicator to form and modify policy. Some policies have used the life-cycle hypothesis, as a guideline to understand saving patterns of different age groups. The hypothesis predicts that individuals will dissave during their youth and older years, and spend middle period saving for retirement. To test this theory, the savings patterns of the elderly have been examined on both the micro and macro level. Unfortunately, both of these analyses differentiate from the life-cycle hypothesis and prove that there are imperfections within it. On the microeconomic level, data about individual households fails to account for intergeneration relations. Continually, though intergenerational relations influences cross-country data, it cannot be used to explain saving decisions at an individual level. Examining the following articles will illustrate the problems with the two sets of data.
Axel Borsch-Supan (1992) examines the consumption and savings of German elderly in the years 1978 and 1983. Following the life-cycle hypothesis, one would argue that elderly would experience dissaving; however, Borsch-Supan’s results contradict this prediction. Unlike the life-cycle hypothesis, which predicts that the elderly will dissave, Borsh-Supan U-shaped graph illustrate that the elderly will save, which implies that theory is wrong. More importantly, though the saving rate is smallest when the elderly first retire, at age 65, the saving rates at any time is at its highest, for individuals over 80 years old. Borsch-Supan gives possible explanations for this discrepancy.
Borsch-Supan lists possible cohort effects to account for the unpredicted results. First, he suggests that income may affect mortality rates, because rich people might live longer than poorer people.[1] Additionally, it is believed by Borsch-Supan that the poor are more likely to enter institutions. Unfortunately, Borsch-Supan’s sample fails to include institutions, which would make it an inaccurate representation of the true population. Continually, Borsch-Supan goes on to analyze the difference in valuation of savings and assets. He examines the valuation of assets at the mean and median levels for individuals. Similar to the omission of the institutional population, Borsch-Supan argues that the difference in valuation of assets will not have an effect on the sample, and the sample tested is a significant representation. Despite these possible cohort effects, Borsch-Supan concludes that they do not have a substantial effect on the savings patterns of his sample.
Following Borsh-Supan discussion of cohort effects, he goes to examine possible economic explanations for the increase in savings at old age. It is possible for the health of the elderly to decline, which in turn limits their consumption habits.[2] Borsch-Supan calls the limited spending habits of the elderly “consumption constraints.” He cites examples such as, the elderly spending less for food and being less mobile to spend time traveling. Additionally, the elderly are more risk-adverse than the younger generations, because they fear not having the means to pay for unanticipated health deterioration. All of these are possible explanations to the increase in the savings rate of the elderly, in Germany.
Regardless of the reasons for the increased saving habits of the German elderly, they are still consuming less then they are saving. Considering this conclusion, Borsch-Supan argument alludes to the fact that elderly may be receiving too much annuity income then what is necessary to survive. Continually, Germany needs social security reform, like reducing the amount of annuity the elderly receive. By confronting the inefficiency in the current system, Germany can improve their policies that grant too much relief to its elders.
David Weil (1994) examined the relationship between the age and savings, and he found a variation between his results on the macro and micro levels. His aggregate analysis concluded that the elderly are not saving, contrary to the individual level where data depicted elders saving for the future. Weil suggests that intergenerational relations, as a possible explanation to these problems. Specifically, bequests to younger generations could influence them to increase their spending habits, while the older generations will save more in anticipation of leaving inheritances. Weil tries to account for the discrepancy, by the examining the effect of intergenerational relations on his results.
Weil argues that the discrepancy between the savings of the elderly at a household level and at a national level would be accounted for, if the survey included intergenerational relations. Intergenerational relations affected the micro results, because they examine a single household, and assume individual households will not influence other households. Weil states,
One implication of this analysis is that, if intergenerational relations are important, one cannot use the mean saving of people at different ages (or any other coefficients that come from micro data that do not account for member of other generations) to forecast changes in the aggregate saving rate in response to changes in the age structure of the population.[3]
Unfortunately, the savings rate of older households could cause younger households to save less, which explains why intergenerational relationships should be accounted for in all surveys.
Weil analyzes the effects of bequests on savings to households that have received inheritances and on households that expect an inheritance. To examine the effects of bequests on savings, Weil evaluates the results from the 1984 Panel of Study Income Dynamics. The survey data includes information about households’ assets and whether the household has received or expects to receive a bequest in the next ten years. The survey runs regression analysis on its data, and the results prove, “families that expect a bequest consume 4.8 percent more than families that do not, and that families that have received a bequest consume 10.4 percent more than those that have not.”[4] The results indicate that the savings of the elderly do affect the savings of others, and which helps explain the difference between the macro and micro results.
Weil concluded that intergenerational relations will affect savings habits of individual households, and that economists should account for these imperfections in data, while making conclusions. Specifically, Weil argued that bequests are a possible motive to influence the spending of younger generations. The results are significant to prove that the micro data cannot be used to predict savings across generations. Additionally, the macro data is insufficient in testing the life-cycle hypothesis prediction about individual behavior, because its data cannot distinguish characteristics on an individual level. Despite this, aggregate data can be used to predict demographic changes in society, where as the micro data cannot.
Richard Disney’s article, “Ageing and Saving”, compares the accuracy of the life-cycle hypothesis to the macroeconomic and microeconomic effects on aging. Like Weil, Disney acknowledges the differences between data on an individual and cross-country level. Unlike Weil, he analyzes data that compares the relationship between the age dependency ratio to the household saving rate and the growth rate, within 19 Organization for Economic Co-Operation and Development (OECD) countries. Using the life-cycle hypothesis, Disney attempts to predict the affect of age dependency ratio to the growth of countries in the year 2025. Continually, Disney examines microeconomic evidence, where the savings in individual households has increased. Similar to Weil, Disney cites intergeneration relations and fear among the elders, as possible explanations to this effect.
United Nations predicts that age dependency ratio will double in the next thirty years. The age dependency ratio is the proportion of people over 65 to the number of people, between the ages of 15 and 65. According to the life-cycle hypothesis, this increase should not affect the relationship between the age dependency ration and the household savings rate and the growth rate, but the relationship should remain negative.[5] Disney’s data from 19 OECD countries in 1977-92 coincides with this prediction, as there is a slightly negative relationship between the age structure and the household savings rate. Continually, 1977-92, Disney compared the age dependency ratio with the average GDP growth. Again, there was also a negative slope, which the hypothesis would predict. Similarly to the results in 1977-92, Disney expected results of the age dependency ration on household savings and the growth rate in 2025, also live up to the life-cycle hypothesis predictions.
Though the cross-country data adheres to its predictions, the microeconomic data does not meet the life-cycle hypothesis expectations, because it fails to account for the true savings behavior of the elderly. Disney cites Weil’s (1994) intergenerational relations as a possible problem for the increased savings on an individual level. Younger generations who expect inheritances will in turn curb their savings habits. He also adds that the elderly will save more, because there is uncertainty of life expectancy, and the fear for the future generations.[6] Though Disney’s cross-country data meets expectations, microeconomic data gives different results, and the discrepencies in the data warrant further emphasis in understanding the individual relationships of the elderly to younger generations.
The savings of the elderly affects national savings, and can indicate modifications to policy. The discussion of the previous three works illustrates the problems interpreting macro and micro data. On the aggregate level, the macro economic data validates the life-cycle hypothesis, because its data indicates that the elderly dissave. The macro data is accurate, because it accounts for intergenerational relationships and other factors, which are omitted at an individual level. On the other hand, the Borsch-Supan’s article indicates that on an individual level the elderly in Germany are saving more, which prove imperfection of the life-cycle hypothesis. These articles prove that policy cannot be formed without understanding the different implications at the micro and macro levels. The implications on national savings of the data at the micro and macro levels should be taken into account, before modifying policy.
Borsch-Supan, Axel. “Saving and Consumption Patterns of the Elderly: the German Case”, Journal of Population Economics, Vol. 5, No. 4, November 1992.
Disney, Richard. “Ageing and Saving”, Fiscal Studies, 1996, Vol. 17, No. 2.
Weil, David. 1994, “The Saving of the Elderly in Micro and Macro Data.“ Quarterly Journal of Economics. February, Vol. 109, No. 1.
[2] P298. Ibid
[3] P67. David Weil., 1994, “The Saving of the Elderly in Micro and Macro Data.“ Quarterly Journal of Economics. February, Vol. 109, No. 1.
[4] 73. Ibid.
[5] Richard Disney. “Ageing and Saving”, Fiscal Studies, 1996, Vol. 17, No. 2. (89).
[6] Ibid. (99).