About the Human Capital Score™

The Human Capital Score™ is a proprietary credit risk measure developed by People Capital. It calculates future income potential by including variables such as GPA, standardized test scores, college and major. For students - who have short or no credit history - a traditional credit measure is unlikely to be an appropriate measure of credit risk.

Frequently Asked Questions (Human Capital Score™)

What is the Human Capital Score?

The Human Capital Score™ (HCS) is a proprietary credit risk measure developed by People Capital. It calculates future income potential by including variables such as GPA, standardized test scores, college and major. For students — who have short or no credit history — the FICO® Score is unlikely to be an appropriate measure of credit risk. While the current version of the Human Capital Score™ is best calibrated for Bachelor's degrees, we are working on future enhancements that will make the HCS applicable to other degree types.

Back to top

Why hasn't this been done before?

Traditional credit scores (such as FICO® Score or VantageScore) are based on attributes that are easy to quantify and rank. More delinquencies lower the score; while a longer, or more positive, credit history raises the score. Alternatively, People Capital uses student attributes to calculate the Human Capital Score™. However, evaluating student academic attributes is not so simple. How do we know which schools or majors are more likely to correlate with the ability to earn income and the capacity to repay a loan? We must collect, clean, and integrate additional data about schools, majors and college students. This requires expertise, time and dedicated resources.

Back to top

What insights does the Human Capital Score™ provide?

The Human Capital Score™ (HCS) projects the possible income paths of college students in the 10 years after graduation. This allows the classification of students into various risk categories which lenders can use to consider the capacity of a given group of college students to repay loans of long- and short-maturities. The projected income shortly after graduation is a good indicator of short-term capacity to pay. Longer-term loans can be assessed by looking at predicted income over a longer period. For benchmarking purposes, we provide a Human Capital Score™ for students for the period 2 years and 8 years post-graduation. Note that the current version of the Human Capital Score™ is best calibrated for Bachelor's degrees, we are working on future enhancements that will make the HCS applicable to other degree types.

Projections of the average future income should be considered with full understanding that it is a computer based algorithm that uses historical data and a broad view of future economic conditions to generate a result. As with any score that tries to categorize individual credit capacity, there is a range of possible future results. The Human Capital Score™ offers broad ranking categories as well as measures of each income path's range. This makes it possible to evaluate the likelihood that income will fall below a certain threshold in a given year, or that average income will fall below a certain threshold in the 10 years following graduation.

Back to top

What high schools do you list?

The high schools that are currently listed come from government data on public and private schools in all 50 states, the District of Columbia, and US territories. The current list comes from the 2007-2008 academic year. If your high school is not listed, please check alternative spellings or abbreviated or alternate names. (For example, James Madison High School in Fairfax County, Virginia, is listed as "Madison High" under the state of Virginia).

If you still cannot find your high school, please email info@people2capital.com with information about the institution.

Back to top

What if I have attended several different high schools or participated in alternative high school programs?

When the Human Capital Score™ Calculator asks for high school information, we are interested in the high school that has awarded, or the institution that is expected to award, a high school diploma or equivalent degree to a student. We do not list single semester, college-in-the-schools, or abroad programs in the high school lists. This is due to the fact that, oftentimes, these programs are unable to confer diplomas to their students. Similarly, early graduation and any supplemental community college enrollment during any remaining “traditional” high school years should be entered into the calculator as the school that awarded the high school diploma or equivalent degree. If your secondary education did not result in the conferral of a high school diploma from a secondary education institution, please select the radio button for “Other”.

If you have received, or expect to receive, a diploma from a school not currently listed among the high schools, please email your inquiry to info@people2capital.com with information about the institution.

Back to top

What colleges and universities does the Human Capital Score™ evaluate?

The Human Capital Score™ lists any postsecondary institution that participates in, or is an applicant for participation in, any federal financial assistance program authorized by Title IV of the Higher Education Act of 1965. The current list of institutions consists of over 7,000 postsecondary education institutions from 2008 in the United States, the District of Columbia, and Puerto Rico—including research universities, state colleges and universities, private religious and liberal arts colleges, for-profit institutions, community and technical colleges, and others. There are several colleges in the United States that are not evaluated by the Human Capital Score™; oftentimes, this is due to the fact that the institution either does not accept federal funding or was established after 2008. We currently list only colleges and universities in the 50 United States, the District of Columbia, and Puerto Rico.

If you don't see your school in the list above, please speak with your financial aid office to see if your institution participates in, or has applied for participation in, federal financial assistance programs. If your institution does participate in these programs and you are still unable to find it on our list, please email info@people2capital.com with your inquiry.

Note: Not all schools on the Human Capital Score™ school list qualify for Title IV Federal Funding and students at certain ineligible schools may be unable to register for loans on People Capital's peer-to-peer lending site.

Back to top

Does the Human Capital Score™ evaluate any online bachelor degree programs?

The Human Capital Score™ can evaluate bachelor's degrees earned at online institutions. However, the school must be listed in the current list of colleges and universities. This list includes postsecondary education institutions that utilize, or have applied for, federal financial assistance as of 2008.

Most online degrees are offered by for-profit institutions or community colleges. If you are unsure of the state under which a specific program is listed for the Human Capital Score™, it is best to first look under the state in which the community college district is located or the state in which the corporate headquarters are located. For example, “University of Phoenix-Online” is listed under the state of Arizona because the corporate headquarters for the University of Phoenix are located in Arizona.

If you cannot find the school, please send your inquiry to info@people2capital.com.

Back to top

What majors does the Human Capital Score™ evaluate?

The Human Capital Score™ currently evaluates 63 different majors that range from Nursing to Philosophy. Since every institution has unique names for academic departments, this list is not by any means comprehensive; therefore, the list of majors used by the Human Capital Score™ will not always match your official major. However, the majors listed do span most academic subject areas, so if your major is not listed, there is most likely a major in the list under which your specific concentration would fall. Please note that we also list "Other" and "I don't know", in the instance you are unable to find your concentration the list or majors, or have not yet decided what your major will be.

Back to top

What type of degrees does the Human Capital Score™ evaluate?

The current version of the Human Capital Score™ is best calibrated for Bachelor's degrees; we are working on future enhancements that will make the HCS applicable to other degree types.

Back to top

How does the Human Capital Score™ address economic condition fluctuations?

Traditional credit scoring models are based on historical data about defaults and credit attributes of borrowers (e.g., debt outstanding, number of credit cards, etc.). People with attributes linked to low default rates in this historical data are given high scores; people with attributes linked to high default rates are given low scores. Such traditional models do not seek to understand why, or how, this link exists between an attribute (e.g., number of credit cards) and loan default. Rather, they rely on patterns and links between these credit attributes and loan default from their historical data.

Instead of focusing on factors that predict credit default, the Human Capital Score™ uses both achievement and academic data to project future earnings potential, which may better inform lenders about an individual's future ability to repay a loan. Our model identifies the set of earnings paths (around a projected average) possible for a borrower with a given set of attributes (e.g., major, school, SAT score, GPA, etc.). Given these earnings paths, we can determine how often an individual is likely to be able to generate sufficient income to pay off a loan. Since insufficient income is often a reason for loan defaults, and we model income level, it is easy to adjust the results to projected changes in economic conditions. When income potential falls in response to changing economic conditions, the Human Capital Score™ will reduce income values but the relative credit-worthiness of a student as compared to other students does not change.

The Human Capital Score™ model provides two relevant insights about the earning potential of college students. The first is the actual Human Capital Score™, which gives a student a "score" (from 1-9) ranking their ability to earn income relative to other college students. The second insight provided by the Human Capital Score™ is an actual estimate of the income they are likely to earn in the future, based on the same historical data. The Human Capital Score™ is not affected by economic fluctuations because it ranks students' ability to earn income relative to one another. This flexibility makes the Human Capital Score™ a superior tool for rank ordering students who, as a general matter, have no significant credit history. It reflects that a students' ability to pay is directly dependent on their potential earning capacity within a future economic context.

However, since the income projections estimated by the Human Capital Score™ are actual income values, this portion of the model is sensitive to cyclical changes in the economy (i.e. inflation, unemployment, technology) and, therefore, these predictions must include the effect of economic cycles on earning potential. In addition to inflation, a "recession adjustment" is currently incorporated into the predicted income values provided by the Human Capital Score™ Calculator.

Since the current economic downturn has had a negative effect on the employment prospects of college graduates, income expectations should be adjusted for the subsequent effect of unemployment on graduates' expected incomes. We use the unemployment rate as an instrument for the effect on potential earnings because unemployment is affected in much the same way as college students' earning potential is affected by a recessionary cycle: in order to earn an income, one must be employed, and fewer job prospects reduce the probability of employment. Therefore, we expect that a higher unemployment rate will result in lower expected earnings for a college student.

Since students generally find work in the same state as the postsecondary institution from which they recently graduated, we calculate a "recession weight" derived from the change in each state's unemployment rate between the current year and the years of the historical data used in the model. Since we observe the change in state-specific unemployment rate, we can adjust our income predictions for the current economic downturn while still using the historical data. The income predictions are adjusted in proportion to how much the state-specific unemployment rate in the current year differs from the unemployment rate experienced by the college students in our dataset.

Back to top

On what scale is the Human Capital Score™ provided?

The Human Capital Score™ ranking scale uses 9 categories (1-9) with "+" and "-" to denote scores that are at the higher or lower ends of the category. Currently, we provide a Human Capital Score™ for students for the period 2 years and 8 years post-graduation.

A score is an opinion of the likelihood that a borrower will have the capacity to generate sufficient income at the end of the given period:

9 Highest 6 Above Average 3 Weak
8 Very Good 5 Average 2 Very Weak
7 Good 4 Below Average 1 Lowest

+ indicates at the highest end of the range
- indicates at the lowest end of the range

Back to top

So how does the Human Capital Score™ work?

The Human Capital Score™ combines credit-risk tools and metrics with academic achievement information to generate potential future earnings and give lenders additional insight into each borrower's future creditworthiness.

The model driven calculation is based on statistical data on a large number of students, with information on their majors, schools, grades, scores, and a host of other attributes. We also know how much these students earned in the years after they graduated. We can use this historical information to create projections of income for students based on each student's specific academic attributes. In overly simplistic terms, if students in our data who study engineering and have good grades had high and growing incomes after graduation, the Human Capital Score™ will assign high and growing incomes to engineering students with good grades who ask for a Human Capital Score™.

The Human Capital Score™ also incorporates information on how much students with various majors earn, the attributes of the various schools, etc. This allows the model to make quality projections of the future potential incomes of students by evaluating the relative effect of attributes on overall earning potential, even when we don't have data on many (or even any) students who went to that school or had that major.

Of course, we can't get data on the incomes of students 10 years after graduation, except from students who graduated at least 10 years ago. To ensure that the Human Capital Score™ reflects the most recent patterns in graduate incomes, we consider the most recent trends in the overall income distribution of college graduates and adjust this distribution based on current economic trends. Also, the current version of the Human Capital Score™ is best calibrated for Bachelor's degrees (but we are working on future enhancements that will make the HCS applicable to other degree types).

Because we have individual-level data on many students, we can project both average likely income and the range of possible income paths. Students from a given major and school may all have relatively similar incomes; another major or school may have wide variation in graduates' incomes. The Human Capital Score™ will be able to provide a variety of statistics relevant for repayment, not just expected income. The model can also estimate the likelihood that income will fall below a certain value, or fall in the worst 10 percent group. We can compute the probability that lifetime income will fall below a given threshold.

Note that the current version of the Human Capital Score™ is best calibrated for Bachelor's degrees; we are working on future enhancements that will make the HCS applicable to other degree types.

Back to top

How is the Human Capital Score™ validated?

In developing the Human Capital Score™(HCS) model, the People Capital team has merged data from several institutional and government sources. We use these data about US postsecondary students to predict each student's future earnings for the decade after graduation. We validate the HCS™ model's predictions in two ways in our original sample.

HCS™ Validation Part 1: Testing Income Prediction Accuracy

First, we evaluate the accuracy of HCS™ predicted income by comparing predicted and realized income in predicted income deciles. Figure 1 illustrates how close the HCS income prediction is to real income in year 4 after college graduation. In this figure, our measure of "accuracy" is determined by the proximity of the results to the ideal: that realized income matches the predicted income value so that the decile data points lay on the 45° line.

To illustrate the accuracy of the HCS™ income predictions, we calculate mean predicted income for each individual and sort individuals by decile of predicted income. For each decile, we plot the average predicted income against the average realized income within a given decile. We “normalize” the results by dividing all numbers by the overall sample's mean realized income. We use these fractions instead of actual income values in order to normalize the results so as not only to preserve the integrity of People Capital's intellectual property , but also to make the results comparable from year to year. In Figure 1, each point represents a decile group, the x-axis measures the decile’s predicted income fraction and the y-axis measures the decile's realized income fraction.

Figure 1: Validating the Human Capital Score™ Income Predictions, Year 4


Ideally, all ten points would lie on the 45° line in Figure 1, thus indicating that the realized income and predicted income fractions are equal. As Figure 1 shows, the 5-year income fractions are fairly close to the 45° line, indicating that the average HCS™ income predictions are almost equal to the average realized income for a given decile of predicted income.

HCS™ Validation Part 2: A Simulated “Out-of-Sample” Test with Split Data

We gain additional validation by randomly splitting our data in two, estimating our model on one half of the data and checking its fit in the other half. This is as close to an out-of-sample test as is possible without new data. Such tests are oftentimes employed to evaluate whether or not a model can accurately forecast for an "out-of-sample" dataset (data not used in the creation of the model) and ideally, an entirely new dataset would be used in this test. However, an alternative dataset containing all the necessary information is not available from an external source nor from People Capital’s in-house data collection efforts. Therefore, we must rely on the aggregated dataset used in the creation of the model to populate both the in-sample and out-of-sample datasets necessary for this validation test.

In order to simulate an out of sample test, we split the dataset randomly into two parts: the first half of our initial dataset (the in-sample dataset) is used to generate the model, while the second half of our initial dataset (the out-of-sample dataset) is used to measure the accuracy of the generated model. The simulated "out-of-sample" test evaluates the model's forecasting accuracy because a resulting income prediction may appear to be a "good fit" for in-sample data, but may not be accurate in forecasting income for out-of-sample data. Ideally, the out-of-sample data would result in predicted incomes as accurate (i.e. as close to the corresponding realized income, or 45° line) as the in-sample data. This result would indicate that the model not only fits the in-sample data, but also that it accurately predicts income for new observations not used in the creation of the model.

Figure 2 illustrates the accuracy of the HCS by showing mean 4-year predicted income and mean 4-year realized income for the predicted income deciles for two sets of data points—one set represents the predictions for the in-sample data and another set that represents the predictions for the out-of-sample data.

Figure 2: Out of Sample Test to Validate the Human Capital Score™ Income Predictions


Ideally, both sets of data points would reside on the 45° line, indicating that the resultant "accuracy" of the HCS income predictions (closeness to realized income values) does not differ between the in-sample results and the out-of-sample results. The proximity of both sets data points to the 45° line indicates that both the "in sample" and "out of sample" mean predicted incomes are close to mean realized income for each of the predicted income deciles.

The "out of sample" data points represent the predicted income deciles' aggregate accuracy of the HCS™ 5-year income predictions for the out-of-sample half of the original dataset—which is the closest we have to an "out of sample" dataset. The out-of-sample income predictions are almost as similar to the out-of-sample realized incomes as the in-sample predictions are to the in-sample realized incomes. In terms of validation, these results indicate that the model maintains its accuracy when used with observations outside of the regression sample. This simulated "out of sample" test validates the HCS™ model's ability to forecast income accurately for an out-of-sample student.

Back to top

What can't the Human Capital Score™ do (aka "the fine print")?

While the Human Capital Score™ calculates potential future income, and thus broadly estimates the ability to pay, it does not measure the willingness to pay. If someone with a high income is unwilling to make loan payments, or someone with no income still makes loan payments, this isn't captured by the model. Therefore, the Human Capital Score™ evaluates future ability to pay, not propensity to pay.

While the Human Capital Score™ ranks future income projections for college students in the 10 years after graduation, we do not have a crystal ball. These projections are based on data about the incomes of people who have already graduated from college and are working now. If economic conditions shift in unexpected ways, we won't capture that. Most obviously, if the current recession reduces the incomes of college graduates unpredictably in the coming years, our estimates of income will be systematically too high. That said, it will continue to show the relative ranking of college students; moreover, it will continue to show which students are relatively better "risks" than others.

The Human Capital Score™ can only project income using standardized attributes and ignores specific student interests or plans. An engineering major from MIT with high scores and grades will have a high Human Capital Score™ because past engineers from MIT with high scores and grades have on average enjoyed high incomes after graduation. If this particular student plans to join the circus (no disrespect to this particular career path intended, just that it traditionally affords a lower income level) after graduation, the Human Capital Score™ cannot, and does not attempt to, reflect this. We are not able to reliably verify or validate information about a specific student's work plans or expectations. We can only rely on information that we can verify (such as SAT score or school attended).

Note that the current version of the Human Capital Score™ is best calibrated for Bachelor's degrees; we are working on future enhancements that will make the HCS applicable to other degree types.

The Human Capital Score™ and any income projections and ranges are solely as statements of opinion and should not be construed as statements of fact. Human Capital Scores are not recommendations to buy, sell or hold any security or to lend to any borrower. People Capital relies on information provided by borrowers and, unless specificially noted, performs only limited verification of this information. The use of the Human Capital Score™ should not be construed as an endorsement of the accuracy of any of the data or conclusions, or as an attempt to independently assess or vouch for the financial condition of any borrower.

Back to top

What is the "College Planning Tool" version?

The Human Capital Score™ College Planning Tool is targeted for students who are planning to go to college and need help measuring the economic value of various schools they are considering. Namely, the tool can help students decide whether it is worth the money spent to go to one school as compared to another, based on the income potential from the academic choices they make. The planning tool works when a user inputs key data about themselves (GPA, SAT scores, planned college major) and the various schools they are considering. The tool then calculates the data and presents a graph and chart documenting results of several scenarios (up to five maximum) of the user’s potential income 10 years after graduation, allowing the user to compare the results between colleges she is considering.

This tool can also be used by college planning consultants and high school guidance counselors with a professional version available for their consulting needs.

Back to top

Can I see a sample report?

Sample report for the Human Capital Score.

Sample report for the College Planning Tool version of the Human Capital Score.

Back to top

The platform is now open for registration.

Join now!

The Peer Advantage

For Borrowers
  • Superior methodology for assessing credit risk to facilitate funding — even with no credit history.
  • Better rates compared to credit cards with low limits, near-term pay-back requirements, etc.
  • Non-conventional loan sources packaged into tax-efficient options.
For Lenders
  • Robust functionality to search, filter and match borrowers on a variety of human capital, financial and educational metrics.
  • Create a unique portfolio of select student loans using benchmark income projections and similar data.
  • Human Capital Score&trade provides a superior way to assess borrowers' relative credit risk.
Bookmark and Share