Corollary 3 Marginal Rates of Substitution of the Tax Rates for the Log- Utility Case) Let the utility function of the representative household be logarith-
3.3 Income Brackets and Income Sources
The use of an income variable smoothed of the variability of temporary shocks displays the importance of analyzing the full spectrum of income sources. Such a process may then provide insight as to the discrepancies seen in earnings for those at different weighted percentile brackets within the spectrum of earnings agents in the economy.We begin to focus here on the three primary lower, middle and upper classes in the U.S.
Some of the trends that present themselves in the data set regard the utilization of stocks and bonds for determining asset returns. Additionally, a greater incorporation of business assets accompanies increases in the earnings brackets. For the purpose of exposition it helps to differentiate between financial and non-financial asset components of net worth. Lastly, all asset categories follow the definitions laid forth in the survey of consumer finance code book.
It is worth noting that in SCF data sets, bonds do not include bond funds or savings bonds but rather include state and local tax-exempt bonds, mortgage-backed bonds, U.S. government bonds and bills as well as corporate and foreign bonds.
Additionally, liquid retirement accounts include IRAs, thrift accounts and future pensions.31Other managed assets include trusts, annuities and managed investment with an equity interest.
All abbreviations in Figs.8and9, follow the definitions laid out in the supple- mental SAS code found on the Federal Reserves public website.32 However, for exposition ‘retqliq’ refers to quasi-liquid retirement accounts such as IRAs, thrift
31A full breakdown of the classification of each individual variable, composition of each variable and calculations is included within the SAS code accompanying the SCF 2014 Bulletin release.
32Precise definitions and calculations for abbreviations in Figs.8 and9 are provided atwww.
federalreserve.gov/econresdata/scf/scfindex.htm.
liq cds
nmmf
stocks
bond
retqliq
savbnd cashli
othma othfin
Income Bracket: 1
liq cds nmmf stocks
bond
retqliq
savbnd cashli
othma othfin
Income Bracket: 2
liq cds nmmf
stocks
bond
retqliq
savbnd cashli
othma othfin
Income Bracket: 3
Fig. 8 Financial assets by income percentile
accounts, future pensions and currently received benefits. ‘Cashli’ refers to the cash value of whole lift insurance and ‘othma’ refers to other managed assets. ‘Liq’ refers to liquid assets such as transaction accounts. ‘CDs’ refers to certificates of deposit;
‘othfin’ refers to other financial assets. Finally, ‘nmmf’ refers to directly held mutual funds excluding money market mutual funds.
Figure8 displays the CPI adjusted financial assets for income brackets based upon percentiles of the lowest 50%, middle 40% and top 10% of income earners for the years 1995–2013. Other than the increases in stocks and bonds for those in the top percentile, it is worth noting that there is at first an increase in retirement assets such as pensions (denoted in yellow) prior to a decrease in the third and highest income percentile.
It would appear that in aggregate there is a decreased ownership of the lower paying assets and an increase in the higher returning assets as one moves up the normalized income ranks. This is particularly noted by the shrinking share of CDs and savings bonds as well.
vehic houses
oresre nnresre
bus othnfin Income Bracket: 1
vehic houses
oresre nnresre
bus othnfin Income Bracket: 2
vehic houses
oresre
nnresre
bus
othnfin Income Bracket: 3
Fig. 9 Non-financial assets by income percentile
It should be noted that these findings are in line with the model simulation in Sect.2. For example, in Sect.2it was suggested that wealthy households engage in consumption smoothing behaviors. The empirical evidence seems to suggest that such families primarily are interested in riskier high yielding instruments to help satisfy consumption when needed. Similar patterns are noticeable when one looks at non-financial assets. Additionally, both Figs.8 and 9 are based upon average numbers for respective categories.
Figure9displays CPI adjusted non-financial assets for the same income brackets for each percentile. One can observe several obvious trends upon simple exami- nation. For example, at the lowest income percentile, vehicle ownership is a larger portion of non-financial assets. As one moves up the income ladder vehicles become a less significant portion of asset holdings.
Once again, the empirical evidence supports the models simulation in Sect.2.
For example, in Sect.2it was suggested that lower income households are simply aiming at consuming necessary goods such as vehicles and housing. In most of
the United States a car is essential to earning a living. Unfortunately, a car is not typically viewed as an investment while a business affords a substantial income stream to create returns for investors and/or smooth consumption for high net wealth households. However, the most notable trends are those regarding both business and housing ownership.
The portion of business ownership (denoted in purple) is at its largest for the top 10% of income earners. It would appear that successful business owners are able to garner a larger portion of income. Additionally, there is a general increase in real estate holdings that are separate from individual housing. For example, at the lower end of the income spectrum, housing constitutes a dramatically large percentage of non-financial asset holdings, yet this tends to decrease with increases in income percentiles.
Once again, for exposition, ‘houses’ refers to primary residence. ‘Vehic’ refers to autos, motor homes, RVs, airplances and boats. ‘Oresre’ refers to other residential realestate such as properties other than principal. ‘Nnresre’ refers to net equity in nonresidential realestate. ‘Othnfin’ refers to other non-financial assets. Finally, ‘bus’
refers to business interests. Once again, as in Fig.8, all abbreviations follow the definitions laid out in the SCF codebook.
It should be noted that other non residential real estate included land con- tracts/notes, properties other than primary occupied residences, time shares and vacation homes. Interestingly, this variable (denoted by “oresre” in the pie charts) tends to increase with income brackets. It makes intuitive sense that as families earn more income they look towards having secondary home options. Clearly this variable is smallest for the lowest 50% of the income earning population in the U.S.
The findings in Figs.8 and9 are for aggregate sample data. It is worth noting that the same findings persisted within the data for individual years. For example, the same trends were present in each individual survey year from 1995 to 2013. It appears that this trend is very persistent in the United States for the income brackets under consideration.