Monday, April 13, 2020

Odds Ratios for Impact of Medical Debt on Home Equity

Odds Ratios for Impact of Medical 
Debt on Home Equity

Question:   Information on people with and without medical debt and people with and without home equity is presented in the contingency table below.   Use this date to calculate the odds ratio for the existence of medical debt on negative home equity.

Follow up to Calculation of Odds Ratio Calculation:   The respondent level data obtainable from the 2018 Financial Capability Survey can be used to estimate a logistical regression model of the impact of medical debt on negative house equity.  What is the odds ratio obtained from this logistical regression model where the dependent variable is a dummy variable set to 1 if the respondent has negative home equity and the explanatory variable is a dummy variable set to 1 if the respondent has medical debt?

Follow up discussion of impact of medical debt on negative house equity:  What do the results presented here indicate about the impact of medical debt on negative house equity?

Contingency Table for Medical Debt
 and Home Equity
Positive House Equity
Negative House Equity
No Medical Debt
Medical Debt
Sample of all homeowners in 2018 track of the national capability study. 

The odds ratio calculation from the contingency table:

Step One:  Calculate the probability of having medical conditional on negative house equity and the probability of have medical debt conditional on positive house equity.

P(medical debt/negative house equity) = 772/1228 = 0.629.

P(medical debt/positive house equity) = 2105/15280 = 0.138.

Step Two:  Calculated the odd for the two conditional ratios above.   

O1= 0.629/(1-0.629) = 1.6929

O2= 0.138/(1-0.13) = 0.1597.

Step Three:  Calculate the odds ratio by taking the ratio of the two odds, which is 1.6929/0.1597 or 10.6.

Follow up to Calculation of Odds Ratio Calculation:

I used survey questions to generate dummy variables for existence of negative house equity and existence of medical debt.  The analysis is limited to homeowners on the 2018 survey.

The code used to estimate the logistical regression model is presented below.

logistic underwater medical_debt if track==2018 & home_owner==1

The results are presented below.

Odds Ratio
Std. Err.
[95% Conf.

Note the odds ratio from the logistical regression model is 10.6 same as from the contingency table.

What is the impact of medical debt on negative equity?

Medical debt has a huge impact on negative equity.  It is likely that people with medical bills are taking out home equity lines or refinancing their mortgages.  Very large implications for retirement security.

I have a larger paper on medical debt and financial distress under review.  Also, please look at my policy primer Defying Magnets: Centrist Policies in a Polarized World.

Sunday, November 17, 2019

Statistics on 401(k) loans and hardship withdrawals

Question: How often in the last 12 months have 401(k) investors obtained a 401(k) loan, made a hardship withdrawal, have either taken out a 401(k) loan or made a hardship withdrawal, or have both taken out a loan and made a hardship withdrawal?

Data:  The data for this question was obtained from the National Finance Capability Study (NFCS) a project of the FINRA Investor Education Foundation.  I examined the two-way frequency table to two questions have you taken out a 401(k) loan and have you made a hardship withdrawal.  People who responded they did not know or did not want to answer were excluded from the analysis.


The two-way contingency table for the two questions is presented below.

Relationship Between 401(k) Loans and Hardship Withdrawals
Withdrawal Y
401(K) Loan Yes
401(K) Loan No

FINRA foundation National Finance Capability Study

The statistics on the incidence of 401(k) loans and hardship withdrawals over the past 12 months are presented below.

Measures of 401(k) Loans and Hardship Withdrawals Past 12 Months
% 401(k) Investors with Loan
% 401(K) Investors with hardship withdrawal
% of 401(k) Investors with either loan or hardship withdrawal
% of 401(k) Investors with both loan or hardship withdrawal

Generated from above table.


·      13.7 percent of households with a 401(k) plan took some money out of their plan in the last year.

·      6.2 percent of the people took out both a 401(k) loan and a hardship withdrawal.

·      Around 73 percent of hardship withdrawals were from people who also have a 401(k) loan.

Policy Analysis:  Many financial planners argue that people should not take out 401(k) loans.   My view is that many of these people have a lot of debt and very little in the way of emergency funds basically need to withdraw some funds when they can’t pay bills.  

My policy recommendation to help these people is to allow them to distribute 25 percent of their 401(k) account without being subject to tax or penalty.   A preliminary version of this proposal is found in the blog post Reinventing the 401(k) Plan.

David Bernstein is an economist who lives in Colorado and is the author of the book Defying Magnets Centrist Policies in a Polarized World.

Note:   Code used to generate table on relationship between 401(k) loans and hardship withdrawals is

tabulate C10_2012 C11_2012 if (C10_2012==1 | C10_2012==2) & (C11_2012==1 | C11_2012==2), cell

Wednesday, September 18, 2019

Modeling House Equity Growth Over Multiple House Purchases

Abstract:   This post shows how to model house equity accumulation over a lifetime for different mortgage choices.   We consider a person who purchases two homes over 25 years, staying in the first home for 10 years and the second home for 15 years.  Two mortgage scenarios are considered.  The first scenario involves two 30-year mortgages.   The second scenario involves two 15-year mortgages.

Question:   How does the choice between a 30-year and 15-year mortgage impact house equity and lifetime debt payments for multiple house purchases?

The Specific Problem:   Our example involves a person purchasing two houses over twenty-five years.   The buyer stays in the first house for 10 years and the second house for 15 years.  

The first house costs $250,000 with the buyer taking out a 90 percent LTV loan.  

The second house costs $500,000 with the buyer placing all funds from the sale of the first house as a down payment on the second house.   The house appreciates at an average annual rate of 1 percent per year.   The seller of the first house pays a 6 percent commission.

The home buyer considers two financing options.   The first financing option involves two 30-year fixed rate mortgage at 3.8 percent.  The second option involves two 15-year fixed rate mortgages at 3.3 percent.   Typically, the interest rate on a 30-year FRM is higher than the interest rate on a 15-year FRM; although, rate levels and the differential between 30-year and 15-year mortgage rates can vary.  The interest rates used in this example are similar to rates currently available in the market.

How do the financing options impact monthly debt payments, lifetime debt payments and accumulated house equity?   


The lifetime debt payment calculations and the accumulated mortgage calculations for the two financing scenarios are presented in the table below.

Two Lifetime House Financing Scenarios

Scenario One
Two 30-year
Scenario Two
Two 15-year 
House One Purchase Price
Loan House One
Interest Rate First House
Years in House
Total House Payments for first House
Appreciation Rate for First Home
Value of First Home on Sale Date
Mortgage Balance on Sale Date
Sale Commission
House Equity from the Sale of First House
Down Payment of Second Home
Value of Second Home
Mortgage on Second Home
Interest Rate on. Second Home
Term (Years)
Years in Home
Total Payments for Second Home
Total Payment on Both Homes
Appreciation Rate for Second Home
Value of Second Home at end of period
Mortgage Balance on Second Home at end of period

Major Differences Between the Two Financing Scenarios:

·      The person who uses two 15-year FRM has a $0 mortgage balance at the end of 25 years after staying in his second home for 15 years.   The person employing two 30-year fixed rate mortgages has 15 years of mortgage payments remaining and an outstanding mortgage balance of nearly $266,000.  

·      The person using the two-15-year FRM paid around $132,000 more in mortgage payments over the 25-year period.  However, the person using 30-year mortgages must either sell the home or continue making payments.   The mortgage is not scheduled to be repaid for 15 more years.

·      The monthly payments on the two 15-year loans are higher than the monthly payments on the two 30-year loans.  The differential is 34 percent for the first home purchase compared to 16 percent for the second home purchase.   The smaller differential in monthly mortgage payments occurs because of the larger accumulation of equity and the larger down payment stemming from the use of the 15-year mortgage.

Comment on Pros and Cons for the Two Financial Scenarios:

Pros associated with use of two 15-year FRM:

·      The advantage of using the two 15-year FRMs is that a home is completely paid for after 25 years.  The person using this strategy can now increase his savings in financial assets if still employed.  The person does not have to disburse funds from a 401(k) plan to pay down a mortgage during retirement.

Cons Associated with use of two 15-year FRMs:

·      The person using two 15-year FRMs makes larger mortgage payments and will probably have to reduce investments in liquid assets to make the larger mortgage payments.

Pros of use of 30-year FRMs:

·      The lower mortgage payments allow this homebuyer to make larger investments in financial assets.

Cons of use of 30-year FRMS:

·      The person after 25 years still has a large mortgage payment.  This reduce future savings in financial assets if the person is still working

·      Additional payments from 401(k) plans to pay down the mortgage are fully taxed as ordinary income. Hence, the person who retires with a 401(k) balance because of the use of 30-year FRMs will have larger tax obligations.  

Caveats on this Analysis:   The potential gain from different mortgage arrangements depends on holding periods for each home, the number of homes purchased during a lifetime, and interest rates.  We look at two different mortgage arrangement for only one scenario.   Results presented for different assumptions on house purchases and interest rates will differ from the results presented here.

Final Question:   Is it better to increase house equity and totally eliminate mortgage debt or should one stay in 30-year mortgages and increase investment in financial assets?  This question is not addressed in this post.  The breakeven point for holding debt must account for taxes when assets are invested in a conventional 401(k) plan because all disbursements from 401(k) plans are taxed as ordinary income.  The desire to avoid financial risk would also favor the mortgage elimination option.  

Certainly, if you are planning to stay in your home after you retire you are exposed to a lot of financial risk if you keep a large mortgage and a large share of your financial assets are equity tied up in a 401(k) plan.  The topic presented here is related to the issue of to what extent investors should prioritize 401(k) saving over other financial priorities.  Go here for a discussion of this topic.

Six Reasons 401(k) plans should not be your top priority