A Guide to the Expenditure Minimisation Problem (EMP)

Imagine you’re shopping with a clear goal: to feel as satisfied as possible while spending the least amount of money. This is essentially what consumers do every day, and economists have developed a powerful framework to understand this behavior-the Expenditure Minimisation Problem (EMP). Unlike the more familiar utility maximization problem where you try to get the most happiness from a fixed budget, the EMP flips the question: how little can you spend to reach a specific level of satisfaction?

The Expenditure Minimisation Problem represents the dual approach to understanding consumer choice. While most of us think about maximizing our happiness given our income, the EMP asks a subtly different question that reveals fascinating insights about how prices affect our purchasing decisions when we want to maintain a constant level of well-being.

Setting up the mathematical framework 🔗

The EMP is formally defined as minimizing total expenditure subject to achieving at least a target utility level. Mathematically, if we have goods with prices and a utility function, we’re trying to find the cheapest combination of goods that gives us our desired satisfaction level.

To solve this problem, economists use the Lagrangian method, which combines the objective (minimizing spending) with the constraint (achieving target utility). The Lagrangian introduces a multiplier that represents the shadow price of utility-essentially, how much an additional unit of satisfaction would cost you at the margin.

The tangency condition 🔗

The solution involves a key condition: at the optimum, the marginal rate of substitution between any two goods must equal their price ratio. This is the same tangency condition we see in utility maximization, which is no coincidence-the two problems are mathematical duals of each other.

Think of it this way: if you’re getting more satisfaction per dollar from apples than from oranges, you should buy more apples and fewer oranges until the satisfaction-per-dollar equalizes. When you’ve minimized your spending to reach a target happiness level, you can’t rearrange your purchases to save money without dropping below that happiness threshold.

Visualizing the solution through diagrams 🔗

The graphical representation of the EMP is elegant and intuitive. On a standard consumption diagram, we draw an indifference curve representing our target utility level. This curve shows all the combinations of goods that give us exactly the satisfaction we’re aiming for.

Next, we consider iso-expenditure lines-these are like budget lines but represent constant spending levels rather than a fixed income. These lines have a slope determined by the price ratio between goods. The solution to the EMP occurs where the lowest possible iso-expenditure line just touches (is tangent to) our target indifference curve.

This tangency point represents the cheapest way to achieve our desired utility. Any other point on the indifference curve would lie on a higher iso-expenditure line, meaning it would cost more to achieve the same satisfaction.

Understanding Hicksian demand functions 🔗

The solution to the EMP gives us what economists call Hicksian demand or compensated demand functions. These functions tell us how much of each good we would buy at different prices while maintaining a constant utility level.

Hicksian demand functions have several important properties that distinguish them from ordinary (Marshallian) demand. First, they are homogeneous of degree zero in prices-if all prices double, the optimal bundle doesn’t change because relative prices stay the same. Second, they are continuous functions of prices and utility, assuming well-behaved preferences.

The no excess utility property 🔗

Perhaps most importantly, Hicksian demand always satisfies what’s called the no excess utility condition. This means the chosen bundle gives you exactly the target utility, not more. Why? Because if you were getting more utility than your target, you could reduce consumption slightly, spend less money, and still meet your goal-contradicting the idea that you minimized expenditure.

Another crucial property is that Hicksian demand always slopes downward with respect to its own price. When the price of a good rises, you’ll always consume less of it while staying on the same indifference curve. This is the pure substitution effect, isolated from any income effects.

Exploring the expenditure function 🔗

The expenditure function maps prices and utility levels to the minimum spending needed. It tells us: “Given these prices, how much money do I need to reach this happiness level?”

This function has beautiful mathematical properties. It’s strictly increasing in utility-reaching higher satisfaction levels costs more money. It’s also increasing in prices-when prices rise, achieving the same utility becomes more expensive. Additionally, the expenditure function is homogeneous of degree one in prices, meaning if all prices double, the minimum expenditure also doubles.

Concavity in prices 🔗

One of the most important properties is that the expenditure function is concave in prices. This reflects consumer substitution behavior: when one price rises, you can partially offset the cost increase by shifting consumption toward relatively cheaper goods. The expenditure function grows with prices, but at a decreasing rate because of this substitution possibility.

Imagine fixing your consumption bundle and calculating what it costs as prices change-this gives you a linear pseudo-expenditure function. But since you can reoptimize your bundle as prices change, your actual minimum expenditure lies below this linear function, creating the concave shape.

Shephard’s Lemma and its applications 🔗

One of the most powerful results in consumer theory is Shephard’s Lemma, which states that the derivative of the expenditure function with respect to any price equals the Hicksian demand for that good.

Why is this true? Consider a small increase in the price of coffee. If you’re currently buying five cups per day, your spending increases by roughly five times the price change. The “roughly” becomes exact in the limit because you’re already optimizing-small adjustments to your consumption have negligible effects on total spending at the margin.

The substitution effect is always negative 🔗

Shephard’s Lemma helps us prove that the substitution effect is always negative. When a good’s price rises and we compensate you to stay on the same indifference curve, you’ll always consume less of that good. This is because the second derivative of the expenditure function with respect to price (which equals the slope of Hicksian demand by Shephard’s Lemma) must be negative due to the expenditure function’s concavity.

This gives us the foundation for understanding consumer responses to price changes. When coffee becomes more expensive, the pure substitution effect-holding your well-being constant-always reduces coffee consumption. Whether you actually buy less coffee in practice depends also on the income effect, but the substitution component always works in the direction of reduced consumption.

Symmetric cross-substitution effects 🔗

Another remarkable implication is that cross-substitution effects are symmetric. The effect of a tea price change on compensated coffee demand equals the effect of a coffee price change on compensated tea demand. This symmetry comes from the fact that both cross-effects equal the same mixed partial derivative of the expenditure function, and the order of differentiation doesn’t matter.

This symmetry constraint is powerful in empirical work. If you estimate how tea prices affect coffee demand, you automatically know how coffee prices affect tea demand-they must be equal when we hold utility constant.

Connecting theory to real decisions 🔗

While the EMP might seem abstract, it helps us understand practical policy questions. When gasoline prices rise, how much do we need to compensate commuters to maintain their standard of living? The expenditure function gives us the answer-it’s the compensating variation that measures the cost of the price change while holding utility constant.

The framework also helps businesses understand demand responses. When you raise your price, some customers leave not just because your product is more expensive, but because their real purchasing power has fallen. The Hicksian approach isolates the pure price effect, showing what would happen if customers were compensated for the income loss.

Consider minimum wage policy: when wages rise, employment might fall not just because labor is more expensive (substitution effect) but also because the firm’s real costs have increased (income effect). The Hicksian framework, applied to firms through cost minimization, separates these effects and helps predict actual responses.

What do you think? How might understanding the difference between substitution and income effects change the way you think about your own purchasing decisions? Can you identify situations where you’ve adjusted your consumption purely due to relative price changes versus changes in your overall budget?

References

1. https://en.wikipedia.org/wiki/Expenditure_minimization_problem
2. https://en.wikipedia.org/wiki/Hicksian_demand_function
3. http://www.econ.ucla.edu/sboard/teaching/econ11_09/econ11_09_lecture4.pdf
4. https://en.wikipedia.org/wiki/Shephard's_lemma