Financial Management and Strategic Management · Treasury and Cash Management

Cash Management Models: Baumol and Miller-Orr

Updated 4 October 2026 · Fact-checked

Cash management models tell you how much cash a firm should hold. The Baumol model gives the optimal transaction size, C = √(2UP ÷ S), when cash use is steady. The Miller-Orr model gives a lower limit, return point and upper limit when daily cash flows are random. Compute them step by step.

Understand Cash Management Models: Baumol and Miller-Orr

A firm holds cash for day-to-day payments, but idle cash earns nothing. If it holds too little, it must sell securities again and again and pays a fee each time. The cash models find the balance between these two costs.

The Baumol model treats cash like inventory, so it is called the EOQ-type model. You start the period with cash, spend it at a steady rate until it hits zero, then refill it by selling securities. A big refill means few transactions but a high average balance and more interest lost. A small refill means more transactions and higher fees. The best refill size is where the two costs are equal.

Baumol works only under certain assumptions: the total cash need is known, cash is used at a constant rate, there are no cash receipts during the period, and the cost per transaction and the interest rate stay constant. Real cash flows are rarely this smooth. That is why Miller and Orr built a second model.

The Miller-Orr model allows daily cash flows to be random. It sets a lower limit (set by management, as a safety cushion), an upper limit and a return point. While cash stays between the limits, you do nothing. If it touches the upper limit, you buy securities to bring cash back to the return point. If it touches the lower limit, you sell securities to bring it back to the return point.

The spread between the limits grows with the transaction cost and with the variability of daily cash flows. It shrinks when the interest rate is higher. So an unpredictable business keeps wider limits, and a business with a high opportunity cost keeps tighter ones.

Key rules to remember

Baumol optimal cash conversion size: C = √(2 × U × P ÷ S); U = total cash needed in the period, P = fixed cost per transaction, S = opportunity cost per rupee for the same period (annual rate for annual U).
Baumol total cost: Total cost = (C ÷ 2) × S + (U ÷ C) × P; First term is holding (opportunity) cost on average cash C ÷ 2. Second term is transaction cost. At the optimum, the two terms are equal.
Number of transactions and average cash: Number of transactions = U ÷ C; Average cash balance = C ÷ 2; Use the optimal C. The period of the interest rate must match the period of U.
Miller-Orr spread: Spread = 3 × [ (3 ÷ 4) × b × σ² ÷ i ] ^ (1/3); b = cost per transaction, σ² = variance of daily net cash flows, i = interest rate per day. If standard deviation is given, square it first. Take the cube root of the bracket.
Miller-Orr limits: Upper limit = Lower limit + Spread; Return point = Lower limit + Spread ÷ 3; The lower limit is given by management. Return point is one-third of the spread above the lower limit.
Miller-Orr average cash balance: Average cash balance = Lower limit + (4 ÷ 9) × Spread; This is the form commonly used in ICAI-style numericals.
Miller-Orr action rule: At upper limit: buy securities worth (Upper − Return point). At lower limit: sell securities worth (Return point − Lower).; Between the limits, no transaction is made.

How to solve Cash Management Models: Baumol and Miller-Orr questions

Decide first which model the question wants. Steady, known cash use points to Baumol. Random or variable daily flows, with a variance or standard deviation, point to Miller-Orr.

1Identify the model from the data: total cash need and a fixed cost per transaction mean Baumol; variance or standard deviation of daily cash flows and a lower limit mean Miller-Orr.
2Write down each symbol with its value: U, P, S for Baumol; b, σ², i, lower limit for Miller-Orr.
3Make the time units match. Convert an annual interest rate to a daily rate for Miller-Orr (divide by 360 or 365, as the question states). Keep U and S on the same period for Baumol.
4Substitute into the formula and show the working line by line. Step marks are given for the substitution.
5For Baumol, find C, then the number of transactions U ÷ C, the average cash C ÷ 2 and, if asked, the total cost.
6For Miller-Orr, find the spread, then the upper limit and the return point. Find the average balance if asked.
7State the decision rule in words: when to buy or sell securities and by how much.
8Write the final answer with units, and add one line of interpretation if the question asks for comments.

Quickest way: Fast route for Baumol and Miller-Orr numericals

When to use it: Use this in the exam when the numerical looks lengthy. It suits both MCQs and the written answer.

Baumol: compute 2 × U × P first, divide by S, then take the square root. Check by testing whether C² × S equals 2UP.
Check Baumol with the equal-cost test: holding cost (C ÷ 2 × S) must equal transaction cost (U ÷ C × P). If they differ, your C is wrong.
Miller-Orr: calculate the bracket (3 ÷ 4) × b × σ² ÷ i in one go. Look for a round number, since exam data usually makes it a perfect cube such as 1,000 or 8,000.
Multiply the cube root by 3 to get the spread. Then add: return point = lower + spread ÷ 3, upper = lower + spread.
In MCQs, remove options that break the rule: the return point must lie between the limits and sit one-third of the spread above the lower limit. A Baumol answer must give equal holding and transaction costs.
In written answers, use this layout: given data, formula, substitution, result, interpretation. This earns step marks even if the final arithmetic slips.

Common mistakes in Cash Management Models: Baumol and Miller-Orr

Using an annual interest rate in the Miller-Orr formula.
The question gives the rate per annum and students plug it in directly.
Fix: Miller-Orr needs the interest rate per day because σ² is daily. Divide the annual rate by the number of days stated in the question.
Using standard deviation instead of variance in the Miller-Orr formula.
The question gives σ and students forget the formula contains σ².
Fix: Square the standard deviation first. If the question gives variance, use it as it is.
Forgetting to multiply the cube root by 3 for the spread.
Students stop after computing the bracket to the power one-third.
Fix: Spread = 3 × cube root. Write the formula with the 3 outside, and check that the spread is larger than the cube root.
Taking the return point as the midpoint of the limits.
It feels natural for the return point to be central.
Fix: Return point = Lower limit + Spread ÷ 3. It sits closer to the lower limit, not at the middle.
Mismatching periods in Baumol, such as monthly U with an annual rate.
Data is given in mixed units and students do not convert.
Fix: Convert so U and S cover the same period before substituting. Annual U goes with the annual rate.
Computing total cost using C instead of C ÷ 2 for the holding cost.
Students forget that cash falls steadily from C to zero, so the average balance is half of C.
Fix: Holding cost = (C ÷ 2) × S. Then verify that it equals the transaction cost at the optimum.

Worked examples

Example 1

A firm needs ₹12,00,000 of cash in a year, spent evenly. Each sale of securities to raise cash costs ₹150. Securities earn 10% per annum. Using the Baumol model, find the optimal cash conversion size, the number of transactions, and the total annual cost.

Show the solution

U = ₹12,00,000; P = ₹150; S = 10% = 0.10.
C = √(2 × 12,00,000 × 150 ÷ 0.10).
2 × 12,00,000 × 150 = 36,00,00,000. Dividing by 0.10 gives 360,00,00,000 (₹360 crore), which is 3,600,000,000.
C = √3,600,000,000 = ₹60,000.
Number of transactions = 12,00,000 ÷ 60,000 = 20.
Average cash balance = 60,000 ÷ 2 = ₹30,000.
Holding cost = 30,000 × 0.10 = ₹3,000.
Transaction cost = 20 × 150 = ₹3,000.
Total cost = 3,000 + 3,000 = ₹6,000. Both costs are equal, which confirms the optimum.

Answer: Optimal conversion size is ₹60,000. The firm makes 20 transactions a year, and the total annual cost is ₹6,000.

Example 2

A company uses the Miller-Orr model. Transaction cost per securities trade is ₹250. The variance of daily net cash flows is ₹16,00,000. The interest rate is 0.03% per day. Management sets the lower limit at ₹20,000. Find the spread, upper limit, return point and average cash balance, and state the action at the limits.

Show the solution

b = ₹250; σ² = ₹16,00,000; i = 0.0003 per day; Lower limit = ₹20,000.
Bracket = (3 ÷ 4) × 250 × 16,00,000 ÷ 0.0003.
(3 ÷ 4) × 250 = 187.5. Then 187.5 × 16,00,000 = 30,00,00,000 (3,00,000,000).
Divide by 0.0003: 3,00,000,000 ÷ 0.0003 = 1,000,000,000,000 (10¹²).
Cube root of 10¹² = 10,000.
Spread = 3 × 10,000 = ₹30,000.
Upper limit = 20,000 + 30,000 = ₹50,000.
Return point = 20,000 + 30,000 ÷ 3 = 20,000 + 10,000 = ₹30,000.
Average cash balance = 20,000 + (4 ÷ 9) × 30,000 = 20,000 + 13,333 = about ₹33,333.
Action: if cash reaches ₹50,000, buy securities worth ₹20,000 (50,000 − 30,000). If cash falls to ₹20,000, sell securities worth ₹10,000 (30,000 − 20,000).

Answer: Spread ₹30,000; upper limit ₹50,000; return point ₹30,000; average cash balance about ₹33,333. At the upper limit buy securities worth ₹20,000; at the lower limit sell securities worth ₹10,000.

Exam tips

Read the data first to choose the model. A given variance or standard deviation of daily cash flows means Miller-Orr. A total annual cash need with a fixed cost per transaction means Baumol.
In MCQs, check the return point, upper limit and spread against the one-third rule. Wrong options often break it.
Questions often ask you to compare the two models. Note that Baumol assumes certain, steady cash use, while Miller-Orr allows random flows and gives a range, not a single size.
Show the formula, the substitution and the unit conversion in the written answer. Examiners award marks for each of these steps.
Add one line of interpretation, such as what the firm does when cash touches a limit. This is often a separate mark.

Practice questions from Treasury and Cash Management

Cash Management Models: Baumol and Miller-Orr in other exams

The same ground in other exams, if you are preparing for more than one or want another angle on it.

Cash Management Models: Baumol and Miller-Orr: frequently asked questions

What is the difference between the Baumol model and the Miller-Orr model?

Baumol assumes cash is used at a steady, known rate and finds one optimal conversion size that balances holding cost and transaction cost. Miller-Orr allows daily cash flows to vary randomly and gives a lower limit, return point and upper limit. Baumol suits predictable businesses; Miller-Orr suits unpredictable ones.

What is the Baumol model formula for CA Intermediate?

C = √(2UP ÷ S), where U is total cash needed in the period, P is the fixed cost per transaction and S is the opportunity cost per rupee for that period. Number of transactions is U ÷ C and average cash is C ÷ 2.

Who decides the lower limit in the Miller-Orr model?

Management sets it, based on the minimum safety cushion it wants against running out of cash. The model then calculates the spread and sets the return point and upper limit from it.

Should I use the daily or annual interest rate in Miller-Orr?

Use the daily rate, because the variance is of daily cash flows. Convert an annual rate by dividing by the number of days the question tells you to use, usually 360 or 365.