Learnings from the lab: Making investing attractive for low- to moderate-income households

Background

Research shows that lower income households are less likely to participate in wealth-generating behaviors such as investing. While income may be part of the explanation for this investment gap, it doesn’t paint the entire picture. Given the potential that investments have for long term returns, there is a need to understand the other factors (e.g., emotional, social, psychological) that drive low- to moderate-income households (LMI) to participate in or refrain from investing.

Key Insights

We conducted online studies to better understand the factors that predict investing behavior among LMI households. In doing so, we examined two hypothesized mechanisms: risk and mental accounting.

Risk: Research by Barauh and Parikh indicates that risk-averse individuals are less likely to invest, and that this outcome varies depending on factors such as gender, age, and financial literacy. However, limited research has been done to understand the nature of this perceived riskiness, and how it varies by income level. LMI households may simply be more sensitive to investment risks because they are more risk-averse than higher-income households. LMI households may not understand the complexity of investing and may see investments as something beyond their financial capabilities. Therefore, we hypothesized that LMI households would perceive investment accounts as riskier than high-income groups would.

Mental accounting: People are more likely to spend money differently depending on where the money is coming from and where it’s going. We refer to this phenomenon as “mental accounting”. We hypothesized that LMI households have preconceived notions about the terms “investing” and “savings” and would therefore be less likely to invest than higher-income groups.

Experiment

We ran two online studies to understand how investing behaviors and perceptions of investing differ among various income groups: low (<$30,000), moderate ($30,000 to $48,000), middle ($48,000 to $73,000), and high (>$73,000).

Study 1 explores the relationships between income levels (low, moderate, middle, and high), financial literacy, types of investment accounts owned, perception of investment risks, and individual factors such as risk propensity and demographics.

Study 2 aims to understand the factors that impact intentions to invest. We examined how risk propensity and mental accounting impact investment decisions using a hypothetical scenario where participants read about an account labeled as investing/savings that had either no, low, medium or high risk. Participants then indicated their choice to save/invest in the account, amount willing to save/ invest, length of time of leaving money in the account, and likelihood of recommending the account to family.

Results

The results from Study 1 indicate that while majority of participants have financial products such as stocks, bonds, and retirement accounts, fewer LMI individuals tend to have such accounts. For instance, while approximately 45% of middle- and high-income people in our sample reported owning stocks, only 11% of low-income individuals do. We also found a positive association between risk propensity and one’s perception of investment risks. In other words, an individual’s decision to invest in a risky investment account was partially accounted for by their perception of whether they are generally a risky person or not. We didn’t find any associations between risk propensity (averseness) and income levels, even after controlling for all demographics. We also didn’t find any associations between income levels and perceptions of investment risks.

These findings suggest that one’s choice to invest in either low or high-risk accounts doesn’t vary by income. However, investment accounts typically have more varied risk levels. We therefore used Study 2 to examine multiple risk options as well as the mental accounting mechanism.

Study 2 was run before the pandemic. We found that account risk (e.g., chance of losing one’s return) had the greatest impact on one’s investment/savings decisions. Moreover, even though riskier accounts had larger associated returns, individuals seemed to be most focused on and most influenced by the possibility of losing all or part of their money. Consequently, participants were significantly more likely to opt for accounts that had lower risk and smaller returns, as opposed to accounts with higher risk and larger returns.

Individual risk propensity also played a role – as the risk associated with the account increased, those with a high-risk propensity were significantly more likely to opt for the higher risk accounts than those with a low-risk propensity. Participants put more money in lower risk accounts and left that money for longer amounts of time. Similarly, people were more likely to recommend lower risk accounts to their family and friends. The investment/savings label did not seem to affect any of these outcomes. Moreover, none of these findings varied by income. We hypothesize that the lack of significant differences may be due to methodological concerns (e.g., hypothetical investment account description didn’t match people’s mental models of investing, people may have been thinking about spending their refund (windfall) in a nonrealistic manner, etc.).

Overall, these findings suggest that risk matters the most; the higher the risk, the lower the desire to put money in the account. Higher risk accounts also lead people to leave less money in the account, and for shorter periods of time, regardless if it’s an “investing” or “savings” account. Given that people are particularly sensitive to risk during trying times, financial service providers that aim to drive savings or investment should be mindful of how they describe their products and the associated risk.

After the start of the pandemic, we reran Study 2 with updated methodologies; we are currently analyzing the data to examine the impacts of the pandemic on these trends