Hello, Please see "project requirements" file for the requirements. I started this project, Please see "my draft" file and need to be completed as the requirements. You need to communicate with me via hangout to discuss any changes in my draft. Thank you and I am looking forward to work and learn from you.
Literature Review From 1983 to 2010, it is estimated that 31% to 53% of working employees had inadequate savings to maintain the same lifestyle during retirement (Benartzi and Thaler, 2013). In addition, 78 million of U.S. employees do not have retirement plans though their employers. The U.S. personal saving rate is the percentage saved from disposable income after paying taxes. From 1975 to 2007, the U.S. personal saving rate declined; however, after the 2008 financial crisis it had a significant rise. In 2013, the U.S. personal saving rate declined again and continued to be low. A higher personal saving rate can have a higher economic growth over the long run. From the household perspective, people save money so they can purchase a house, fund their children's education, protect themselves from emergency expenses, and spend it during retirement. The goal of this paper is to provide a better understanding of the saving rate in the U.S. so that policymakers can design policies to increase the saving rate. These policies can help people to increase their savings and achieve their goals. The life-cycle hypothesis (LCH) assumes that people spend their earnings over their lifetimes. This paper’s focus is on the saving rate in U.S. households in the last decade. Attanasio (1998) analyzes the decline in the saving rate of U.S. households using the Consumer Expenditure Surveys (CE) from 1980 to 1991. Attanasio (1998) finds that the saving rate peaks at the age of 57 as people near retirement. Attanasio (1998) argues that people who were born between 1920 and 1939 had a higher saving rate compared to other cohorts who were born after. Financial literacy plays an important role in preparing for retirement. Lusardi (2007) finds that half of older workers lack a basic understanding of their pension plans. Lusardi (2007) also finds that a large number of workers do not understand the basic rules of the social security benefits. Gunes and Tunc (2018) investigate the effect of mortgage payments on U.S. household personal saving rates over the period of 1999 – 2015. They find that for every 1% increase in mortgage payments, the saving rate decreases by 0.15%. In addition, Gunes and Tunc (2018) believe that government subsidy policies have a negative implication on the personal saving rate which suggests that there is a relationship between monetary policy and saving rate. Sukar and Krishnan (2012) examine the impact of selected variables on the personal savings behavior from 1980 to 2008 and find a negative relation between personal savings and wealth. After adjusting for inflation, Sukar and Krishnan (2012) use six variables consisting of personal savings, net wealth, household debt, government budget balance, and real interest rate and find that the real interest rate has a positive effect on personal savings. Traut-Mattausch and Jonas (2011) hypothesize that there is a relationship between financial satisfaction, income, and saving behavior. As predicted, the relationship is found to be strong. They also compare the relationship between financial satisfaction and saving behavior and find it to be stronger for people with low income than people with high income. Marital status also can affect saving behavior in U.S. households. Knoll et al. (2012) compare the participation rate in a defined contribution pension plan between married and single people between the age 22 – 35 years old and find that married people have a higher participation rate. They also find that single women are poor at saving for their retirement. Cultural differences and backgrounds can affect how people save. Fuchs-Schundeln et al. (2020) find that immigrants tend to save more in Germany and the United Kingdom than the native citizens. Kirsova and Sefton (2007) examine the saving rate based on the life-cycle model of savings and compare the United Kingdom, United States and Italy. They find that the U.S. saving rate is relatively lower than the U.K. and Italy because Americans work more and retire at older ages. The U.S. households have also have lower saving rate than China. Choi et al. (2013) finds that the Chinese households’ income grows faster than U.S. households’ income. As a result of this income growth, the Chinese household saves more than Americans. Examples from those developed countries suggest that the U.S. is doing poorly in savings. Chen et al. (2019) find that the increase of health expenditures affects the personal saving rate. They also find that the development in the health care sector led to a 50% decline in the personal saving rate from 1995– 2009. Mortgage equity withdrawal is when people borrow money against their property. This is also one of the reasons why the personal saving rate has declined in the past years. Chen et al. (2013) investigate the effect of mortgage equity withdrawal on U.S. households over the period 1993–2011. They find a strong relation between saving rate, wealth, and mortgage equity withdrawal. Remble et al. (2014) use the 2007 Consumer Finances survey to understand the saving decisions among small business owners. They find that people who own a family business have a higher probability of saving money every year comparing to non-family business owners. More recently, Ouliaris (2018) develops a time series model to measure consumption and estimate the U.S. personal saving rate. According to that study, the saving rate has increased after the 2008 financial crisis. Ouliaris (2018) tests if this change is because of household behavior, changes but no relationship is found. Ouliaris (2018) explains the change of saving rate because of the change in income, employment, and wealth. Hershfield et al. (2011) suggest an interesting way to encourage people to increase their saving or change their saving behavior. They find that people would save more for retirement when they see age-progressed images of themselves. Outcault (2012) suggests that government entities, financial institutions and employers should work together to create policies to encourage young adults to save or increase their savings. The literature that is reviewed here shows a declining pattern of the U.S. personal saving rate. There is also an indication that older generations of Americans may have prepared more than younger generations for retirement. This paper aims to understand the decline in the U.S. personal saving rate and how prepared are Americans for retirement. Data Section The data that will be used in this paper is from the Consumer Expenditure Surveys (CES). The U.S. Bureau of Labor Statistics (BLS) collects data on expenditures, income, and demographic characteristics of a nationally representative U.S. households. The CES select a sample of participants from different geographic areas. In 2020, 131,234 consumers participated in this survey. The BLS publishes data on consumer expenditures data twice each year by conducting two surveys, the Interview Survey and the Diary Survey. The purpose of the Interview Survey is to collect data about more major recurring expenditures such as rent and utilities. The Diary Survey collects data on frequently recurring expenditures such as food and clothing. After collecting the samples from the U.S. households, the BLS uses those samples to U.S households' expenditures. First, the BLS select a geographic area, and then they select participants in those areas. Each geographic area consists of clusters of counties which is called primary sampling units (PSUs). The BLS divides each of the U.S. into four main geographic regions which is the Northeast, Midwest, South, and West, and each region is divided into further geographic divisions. The BLS selects a random sample of PSUs from each division, and then they select a random sample of U.S. households from each PSU. The CES provide good information about U.S. household income and expenditures and therefore saving. In the paper, the personal saving is the dependent variable, while Region of residence, expenditure expenses, and age are the independent variables. The personal saving rate is the saved percentage of the disposable income after taxes. Income is defined as the total earnings that the household makes from working wages, while expenditures are all the expenses that are related to the household like food, housing, apparel and services, transportation, health care, and entertainment. Model The linear regression model to be estimated is: = + + + + … + + where measures the U.S. household saving and , , … are the explanatory variables which are: Age, Income, Race, Education, Housing, Number of People Living in the Unit, Number of Earners Living in the Unit and Healthcare Expenses is the constant team, through are the slope coefficients for each explanatory variable, and is the error term. The dependent variable represent household saving is used in this paper. Household saving is the difference between income and expenditure. Using on the consumer theory of the life-cycle that hypothesis that people borrow when their income is low and save when income is high. The LCH shows the relationship between income and consumption over the lifetime of household. The main idea of the hypothesis is that people should aim to smooth their consumptions during the different stages in life so they can enjoy having a better lifestyle during retirement. The expected relationships between savings and income, and how age and race impact savings are shown in the following hypothesis: Hypothesis 1: Having a higher income will significantly affect the likelihood of savings. Hypothesis 2: Age will affect the likelihood of saving differently among different age groups. Hypothesis 3: Race will affect the likelihood of saving differently among different race groups. To test whether the consumer theory of the life-cycle and explanatory variables contribute significantly to predict saving, the three hypothesis will be tested. This paper assumes positive assumption on the coefficients of Age, Income, Education, Housing, Number of Earners Living in the Unit, and a positive assumption on the coefficient of Race, Number of People Living in the Unit, Healthcare Expenses in order for the hypothesis to be proven right. Table 1: List of the Dependent Variable and Explanatory Variables Used in the Model Dependent Variable Variable Description The U.S. households saving A continuous Variables for U.S. Households Saving Explanatory Variables Variable Description Expected Sign Age A continuous variable of the Age of reference person + Income A continuous variable for the Income before taxes + Race A Dummy variable of the Race of reference person =1 if Black or African-American =0 if White, Asian, and all other races - Education A Dummy variable of the Education of reference person =1 if completed college =0 if not completed college + Housing A Dummy variable of the housing =1 if Homeowner =0 if Renter - Number of People Living in the Unit A continuous variable for the Average number in consumer unit - Number of Earners Living in the Unit A continuous variable of the Earners + Healthcare Expenses A continuous variable for the Healthcare expenses - References Abdulhamid Sukar, & V Sivarama Krishnan. (2012). The Determinants of Personal Savings in the U.S: The Role of Wealth. Journal of Financial and Economic Practice, 12(3), 36. Attanasio, O. P. (1998). Cohort Analysis of Saving Behavior by U.S. Households. The Journal of Human Resources, 33(3), 575–609. https://doi.org/10.2307/146334 Benartzi, S., & Thaler, R. H. (2013). Behavioral Economics and the Retirement Savings Crisis. Science (American Association for the Advancement of Science), 339(6124), 1152–1153. https://doi.org/10.1126/science.1231320 Caporale,