Savings and mental accounting

Standard economic theory treats money as fungible — a shilling is a shilling, wherever it came from and wherever it's going. Empirically, people don't treat money this way. They put it in mental accounts that have different rules for how it can be spent. Mental accounting (Thaler 1985, 1999) explains a lot of what otherwise looks like irrational saving and spending behaviour.

What mental accounts are

A mental account is a way of categorising and tracking money based on its source, intended use, or storage location. The same shilling has different perceived value depending on its account.

• Income-source accounts — regular salary vs windfall (lottery, bonus, gift). Same shillings, different willingness to spend
• Purpose accounts — food budget, transport budget, schooling budget, entertainment budget
• Storage accounts — wallet cash, M-PESA balance, bank account, SACCO contributions, chama savings
• Time accounts — current month's income, next month's income, year-end bonus, retirement savings

Each account has different spending rules. The food budget is spent on food (even when there's spare money in the entertainment budget). The chama savings are committed to the rotation (even when the household needs the money for something else). The standard model says fungibility should override these labels; reality says the labels often hold.

Where mental accounts come from

Several plausible mechanisms:

• Pre-commitment device — by labelling money for a purpose, you reduce its future flexibility, helping you stick to plans. Mental accounting is partly self-imposed commitment
• Cognitive simplification — tracking total wealth across all transactions is hard. Tracking each account separately (food budget left, transport budget left) is easier even though it's less optimal
• Source-labelling — windfall income FEELS different from regular income, even though the budget effect is identical. The affect-heuristic gives windfalls a 'fun money' quality
• Social signalling — visible accounts (chama contributions, bank balance) signal financial standing; people maintain these accounts partly for signalling reasons

Examples in African households

• Chama membership — savings here are NOT fungible with general household cash. A member won't 'borrow back' their chama contribution for a household emergency, because it's seen as committed to the group
• School-fees savings — many Kenyan households maintain a school-fees fund (often in a SACCO or in a separate bank account) that's effectively unavailable for other uses, even when the household has higher-priority current needs
• Burial and emergency funds — many households have a specific fund for the cost of major funerals (locally significant, often costing KES 100-200 thousand). This fund is sacrosanct
• Livestock-as-savings — owning a goat or a cow has a 'savings' quality. Selling for daily expenses is psychologically harder than withdrawing cash from a bank account, even if the cash value is identical
• Phone-credit balance — small mobile-money balances are sometimes spent more readily than larger amounts in the same account, because they feel like 'play money'

The fungibility violation

Standard economics says: if you have KES 5,000 in your school-fees fund and an emergency, you should use the school-fees fund and replace it later. Mental accounting says: people don't, even when it's clearly the welfare-maximising thing to do.

Empirical evidence (J-PAL Africa and similar): households with separate school-fees accounts demonstrate substantially LESS reallocation across budget categories than those with general-purpose savings. The compartmentalisation is real, persistent, and (in some interpretations) welfare-improving — it helps the household stick to its longer-term financial plan.

Mental accounting + present bias = product design

The most successful behavioural-finance products combine the two:

• M-Shwari Lock Savings — opens a separate mental account ('lock savings') with rules that prevent withdrawal (committing against present bias). The labelling + lock feature reinforces both behavioural mechanisms
• Chama rotation — the chama account has rules (you contribute, you wait your turn, you can't access prematurely) that both segment your spending (mental accounting) and prevent present-bias withdrawal (commitment)
• Goal-based savings — products that name the saving (school-fees, livestock, motorcycle) and tie the savings to that visible goal. Strong mental-accounting reinforcement plus motivational salience
• Layby — pre-paid credit toward a specific purchase. The money is in a 'goods-purchase' account, segregated from general spending

The dark side of mental accounting

Compartmentalisation can hurt as well as help. Examples:

• Burial-fund accumulation — households accumulate substantial savings (often KES 100-200k) for funeral costs. The savings are inaccessible for productive investment, even when the household has higher-return uses for the capital. The cultural-affective importance of proper burial creates the rigid account
• Foreign-aid mental accounting — aid recipients sometimes use aid funds for projects the donor specified (drilling boreholes) when alternative uses (paying school fees) would have higher household welfare. The donor's labelling overrides household-optimal allocation
• Hardship and account-spillover failure — when one account is depleted, households should ideally borrow from another. Empirically, they often don't — they let the depleted-account need go unmet rather than violate the compartment. Children's school fees go unpaid even when chama money is accessible because the household has 'closed' the school-fees account

Implications for SACCO and savings-product design

• Multiple savings buckets within one institution — allow members to maintain separately-labelled savings goals (school fees, asset purchase, emergency fund, dowry). The labels reinforce mental-accounting commitment
• Visualisation tools — apps that show separate progress bars per goal exploit visual salience of separate accounts
• Inter-account transfers — make them possible but require explicit action (not automatic). Explicit transfers preserve the mental-account boundary while allowing genuine reallocation when needed
• Reporting — separate balance statements per account/goal. Reinforces the mental compartment

When to ignore mental accounting in product design

Sometimes the welfare-maximising design fights mental accounting:

• Crisis support — emergency cash transfers should be unconditional precisely because households need fungibility in crisis. The donor labelling 'food assistance' vs 'cash for productive use' is welfare-reducing in crisis contexts; cash gives the family decision-rights
• Investment products — when the goal is asset accumulation toward a portfolio, mental-account separation by holding type (real estate, equity, savings) can lead to suboptimal portfolio construction. Single-portfolio thinking is the correct behaviour but requires deliberate effort
• Inheritance and bequest — the standard economic 'household wealth' view treats wealth as a unified store; mental accounts may compartmentalise wealth in ways that interfere with the household's long-term wealth-management. Estate planning has to navigate this