Narrow and Broad Monetary Aggregates

Why Economists Use More Than One Money Measure

Money is not a single list of objects with a universally fixed boundary. Currency and transaction deposits are highly liquid: they can generally be used directly, or transferred readily, to make payments. Other deposit balances may be accessible but less convenient, or may require a transfer or waiting period. A monetary aggregate is an operational measure that groups specified financial assets together. Narrow and broad aggregates use different boundaries to describe money with different degrees of liquidity.

The distinction is useful because the narrow measure emphasizes assets most readily used for transactions, while the broad measure includes a wider set of monetary assets. Neither measure is simply “the correct amount of money” for every economic question. The result depends on the published definition: which assets are counted, whose holdings are counted, and how the aggregate is constructed.

Definition: A narrow monetary aggregate is a measure focused on the most liquid monetary assets, commonly including currency held by the public and transaction deposits. A broad monetary aggregate includes a wider set of monetary assets, often by adding specified deposit balances to a narrow aggregate. These are general descriptions, not universal component lists: each jurisdiction specifies its own operational definitions.

Liquidity and the Aggregate Boundary

Liquidity describes how readily an asset can be used to make payments or converted into a means of payment, considering factors such as time, cost, and restrictions. Currency and balances in transaction accounts are typically highly liquid. Some other deposit accounts are less directly usable for purchases but may still be included in a broad aggregate under the jurisdiction’s rules.

A broader measure therefore usually covers more than the narrowest transaction-focused measure. In a system where the published broad aggregate contains every component of the narrow aggregate, the relationship can be written as follows. Here, $M_N$ denotes the narrow measure, $M_B$ the broad measure, and $Q$ the additional eligible assets counted in the broad measure but not in the narrow one.

$$ M_B = M_N + Q $$

This formula is valid only when the definitions are nested, the additional components are counted once, and the aggregates use the same reporting date and holder boundary. It is not a license to assume that every measure called “broad money” is calculated in precisely this way. The published definitions must be checked.

Derivation: Suppose the narrow aggregate consists of a set of eligible balances $N$, and the broad aggregate consists of those same balances plus a disjoint set of additional eligible balances $Q$. Adding the balances in $N$ gives $M_N$; adding the balances in $Q$ gives $Q$. Since the sets do not overlap, adding their totals counts every included balance exactly once, so $M_B=M_N+Q$. If all balances are nonnegative, then $Q\geq 0$, which implies $M_B\geq M_N$ for this nested construction. If the two published measures do not have nested definitions, that inequality cannot be assumed from their labels alone.

This also explains the intuition behind the comparison: a broad aggregate adds assets that are generally less immediately usable for transactions, while retaining the assets in the narrow measure in a nested system. “Less liquid” is a relative description, not a claim that the added balances are unavailable or economically irrelevant.

Labels Are Not Universal Definitions

Names such as $M_1$ and $M_2$ are common labels for progressively broader measures in some statistical systems. They are not guarantees that two jurisdictions include identical assets in an aggregate with the same label. One jurisdiction’s definition may count a particular type of deposit in its narrow measure, while another may place comparable balances only in a broader measure or use a different published grouping.

Even within one jurisdiction, it is the current published component definition—not a remembered label—that governs a calculation. The boundaries can reflect the financial instruments and statistical conventions in use. For a careful comparison, identify the actual components before interpreting differences in reported totals. This tutorial uses hypothetical definitions in its examples; none of the figures represents a real economy.

A useful way to read a monetary aggregate is to ask three questions: What asset types are included? Whose holdings count? Is the measure nested within another measure being compared? The first question concerns composition, the second the coverage boundary, and the third whether a direct addition or inequality is valid.

Worked Example: Calculating Nested Narrow and Broad Measures

A hypothetical jurisdiction defines its narrow aggregate as currency held by the public plus transaction deposits. Its broad aggregate includes the entire narrow aggregate, plus eligible savings deposits and short fixed-term deposits. On a reporting date, these balances are $640 million in currency, $360 million in transaction deposits, $420 million in savings deposits, and $80 million in short fixed-term deposits. Calculate both aggregates and the amount added when moving from narrow to broad money.

Method. Apply the stated component definitions, adding each eligible balance once. First calculate the narrow measure:

M_N = 640 + 360 = 1{,}000 \text{ million dollars}

The additional components are savings and short fixed-term deposits:

Q = 420 + 80 = 500 \text{ million dollars}

Because the broad measure includes the narrow measure, add the additional components to it:

M_B = 1{,}000 + 500 = 1{,}500 \text{ million dollars}

Check. Adding all four distinct components directly gives $640+360+420+80=1{,}500$ million dollars, the same broad total. The broad measure exceeds the narrow measure by $1{,}500-1{,}000=500$ million dollars, exactly the amount of the added components. A common mistake is to add the $1,000 million narrow total to a list of broad components that already includes currency and transaction deposits; that would count the narrow components twice.

Levels and Growth Rates Answer Different Questions

A monetary aggregate’s level is its measured amount at a particular date. Its growth rate describes the proportional change between dates. A broad measure can be larger than a narrow measure at both dates while growing more slowly over the interval. Comparing levels alone does not reveal which measure changed faster.

For a value $M_0$ at the start of a period and $M_1$ at its end, the percentage growth over that period is:

\text{Growth rate} = \frac{M_1-M_0}{M_0}\times 100\%

The denominator is the starting value. This calculation gives the simple percentage change for the stated interval; it is not an annualized rate unless the interval and annualization method are specified.

Worked Example: Comparing Growth in Narrow and Broad Money

In a hypothetical economy, the narrow aggregate rises from $800 billion to $880 billion over one year. Over the same year, the broad aggregate rises from $1,600 billion to $1,680 billion. Calculate each growth rate, and state which aggregate grew faster in percentage terms.

Method. Calculate each change relative to its own starting level. For narrow money, the dollar increase is $880-800=80$ billion dollars. Its growth rate is:

\frac{80}{800}\times 100\% = 10\%

For broad money, the dollar increase is $1{,}680-1{,}600=80$ billion dollars. Its growth rate is:

\frac{80}{1{,}600}\times 100\% = 5\%

Conclusion and check. Both measures increase by $80 billion, but narrow money grows faster as a percentage of its initial value: 10% compared with 5%. The broad aggregate is larger at the end of the year, but that does not mean it grew faster. A common mistake is to compare the dollar changes alone and conclude that the two growth rates are equal; the different starting amounts make their percentage changes different.

Why Cross-Jurisdiction Comparisons Need Care

Suppose two jurisdictions publish measures with similar names. A direct comparison of their totals is informative only if the definitions and coverage are sufficiently aligned for the question being asked. If the included assets differ, a higher total could reflect a broader component boundary rather than a greater amount of comparable monetary holdings.

The following example shows how to calculate each measure under its own stated rules. The calculations are valid, but they do not establish that the resulting narrow measures are identical concepts.

Worked Example: Reading Similar Labels With Different Definitions

Two hypothetical jurisdictions report a narrow and a broad measure for the same month. Jurisdiction A’s narrow measure is currency of $520 million plus transaction deposits of $480 million. Its broad measure adds $300 million of savings deposits and $200 million of short deposits. Jurisdiction B’s narrow measure is currency of $500 million plus transaction deposits of $700 million. Its broad measure adds $500 million of savings deposits. Calculate each narrow and broad total, then explain what can and cannot be concluded from the totals.

Jurisdiction A. Its narrow total is:

M_{N,A}=520+480=1{,}000 \text{ million dollars}

The assets added in its broad measure total $300+200=500$ million dollars, so:

M_{B,A}=1{,}000+500=1{,}500 \text{ million dollars}

Jurisdiction B. Its narrow total is:

M_{N,B}=500+700=1{,}200 \text{ million dollars}

Its broad measure adds $500 million of savings deposits:

M_{B,B}=1{,}200+500=1{,}700 \text{ million dollars}

Interpretation and check. Under the definitions given, Jurisdiction B reports $200 million more in its narrow measure and $200 million more in its broad measure. Each broad total exceeds its own narrow total by $500 million. However, the added components differ: A includes both savings and short deposits, whereas B’s stated addition is savings deposits. The totals alone do not show that the jurisdictions’ measures have equivalent boundaries or support a like-for-like conclusion about comparable monetary assets. A common mistake is to treat a matching label as proof that the component definitions match.

Common Mistakes and a Reliable Procedure

The main risks are not difficult arithmetic; they are errors in interpreting the boundary. In an exam-style calculation, use the definition provided rather than importing a component list from memory.

Assuming a label fixes the contents. Treat the stated operational definition as controlling, especially when comparing jurisdictions.
Counting nested components twice. If a broad total is defined as including the narrow total, add only the additional components to the narrow total.
Confusing levels with growth. A larger aggregate can have a smaller percentage growth rate; calculate changes relative to each measure’s own initial level.
Assuming every pair of measures is nested. The relation $M_B\geq M_N$ follows from a nested, nonnegative component construction, not from the words “broad” and “narrow” alone.
Ignoring the reporting boundary. A total is meaningful only with its reporting date and the eligible holders specified by the definition.

Key takeaway: Narrow aggregates focus on the most liquid monetary assets; broad aggregates generally add specified, less transaction-ready assets. To calculate or compare them correctly, use the stated component definitions, check whether the measures are nested, and keep the reporting boundary and date consistent.

Check Your Understanding

For each problem, use only the hypothetical definition and data given.

A jurisdiction defines its narrow measure as $450 million of public currency plus $750 million of transaction deposits. Its broad measure adds $300 million of savings balances. Calculate both totals. (Answer: narrow = $1,200 million; broad = $1,500 million.)
A narrow aggregate rises from $900 billion to $945 billion in a year. Calculate its percentage growth. (Answer: 5%.)
A broad aggregate includes a narrow total of $1,400 million plus $250 million in additional eligible balances. A student reports broad money as $1,400 million plus $1,400 million plus $250 million. Identify the error and give the correct total. (Answer: the narrow total was counted twice; broad money = $1,650 million.)
In a hypothetical system, narrow money is $640 billion at the start of a year and $672 billion at the end. Broad money is $1,280 billion at the start and $1,344 billion at the end. Calculate both growth rates and compare them. (Answer: narrow growth = 5%; broad growth = 5%; they grew at the same percentage rate.)
Jurisdiction X’s narrow measure is $800 million and its broad measure is $1,100 million. Jurisdiction Y reports $950 million and $1,250 million under measures with the same labels, but its published definition includes a different set of additional assets. What is the strongest valid conclusion from these figures alone? (Answer: Y’s reported totals are $150 million higher for each label, but the differing definitions prevent assuming the measures are directly comparable.)

What You'll Learn