How modern monetary systems work / personal lessonsLesson 16 of 16
Lesson browserHow modern monetary systems work lessonsPrevious: Exchange rates and monetary constraints
Monetary systems · Lesson 16

Read a monetary argument from the ledgers outward

Combine accounting, institutional rules, behavior, and evidence in one case.

By the end: Audit an expert claim and state which conclusions are established, conditional, or unsupported.

The invented case

Consider a country with a floating currency and mostly domestic-currency public debt. Banks expand mortgage lending. An energy import disruption raises prices. The central bank raises its policy rate while lending reserves to an eligible bank suffering withdrawals. Later, the government issues bonds to finance targeted energy support. A commentator says all four actions are the same thing: printing money that must raise inflation proportionally.

Start by separating transactions. Mortgage origination creates bank loans and deposits under the usual entries. Central-bank lending creates reserves and a debt owed by the bank; it is not a capital injection. Bond issuance swaps private assets before the government spends. A fiscal transfer changes recipients' income and financial claims under the applicable financing sequence. These operations have different counterparties and risks.

Build a dated transaction record

Audit the commentator's claim by defining what would have to be proportional. If a deposit measure grows by 10%, the claim might predict a 10% rise in a consumer-price index over one year. That statement needs a starting date, an ending date, and a specified deposit measure. It also needs assumptions about production and the willingness to hold deposits. Saying only that money was printed leaves those parts unspecified. A reserve increase, a deposit increase, and a treasury payment are different observations. Combining them into one total can count different financial claims at several layers of the same payment system.

Construct an additional paper record with opening commercial-bank assets of 500, customer deposits of 450, and equity of 50. The bank grants a mortgage of 60 with no fees or immediate loss provision in this example. Its assets become 560 because loans increase by 60. Deposits become 510, and equity stays 50. The borrower gains a deposit asset of 60 and mortgage debt of 60. The new purchasing power is real as a payment capacity, but the borrower has not gained 60 of net worth. Its new liability is as large as its new financial asset.

Suppose the borrower buys an existing home for 60 from another customer of the same bank. The borrower's deposit falls by 60, and the seller's rises by 60. Total bank deposits stay at 510. The borrower now holds the house instead of the deposit and still owes the mortgage. The seller exchanged a house for a deposit. This sale transfers an existing asset, rather than producing another home at the instant of payment. Services involved in arranging the transaction can be newly produced, but the house purchase price is not itself a measure of new construction.

Contrast a mortgage used to pay a builder for newly produced housing. Origination can have the same loan and deposit entries, but production, employment, materials use, and capacity differ. If a builder needs scarce imported inputs, some spending can increase imports or costs. If unused labor and materials are available, production may expand. These are conditional mechanisms, not quantities supplied by the loan ledger. Ask what the loan financed before predicting a consumer-price effect. A house-price increase and a consumer-price-index increase also refer to different measured objects, so their percentages cannot be substituted without explanation.

Record the energy disruption separately. Suppose an importer previously bought 20 units of fuel at 2 currency units each and now must pay 3 each. Its bill for the same physical quantity rises from 40 to 60. The extra 20 is a higher payment to the supplier, not evidence that the domestic mortgage entry caused the disruption. The importer might raise selling prices, accept a lower margin, reduce purchases, or combine these responses. The observed cost change establishes a pressure on its accounts. Which response actually occurs needs contracts, quantities, margins, and dates rather than another restatement of the money total.

Move from accounting to mechanisms

Then ask what spending and production do. New mortgages can bid for existing homes, finance construction, or replace other financing. Their consumer-price consequences depend on those uses and the response of supply. An energy disruption can raise an import cost even if domestic credit is weak. A policy-rate increase can restrain later demand while emergency liquidity keeps payments functioning.

Fiscal support may protect vulnerable households but also sustain demand for scarce energy, depending on its design and funding. It can alter measured consumer prices directly if it changes administered prices or subsidies, while leaving underlying resource costs high. These effects require a behavioral and measurement account. The ledger alone does not determine their size or whether the policy is desirable.

Trace separate interventions through their counterparties

A liquidity operation should be judged against a payment obligation. Consider a separate paper bank with reserve assets of 20, loans of 240, and bonds of 40. Its assets total 300. Deposits of 270 and equity of 30 finance them. Customers request a transfer of 35 to another bank. Under immediate reserve settlement without credit, it is short of 15 reserves. A central-bank loan of 25, assuming eligibility and adequate collateral, raises reserves to 45 and adds borrowing of 25. Assets become 325, liabilities become 295, and equity remains 30. Borrowing changes available payment funds, not the bank's residual ownership claim.

After the 35 transfer, this bank has reserves of 10, loans of 240, and bonds of 40, totaling 290. Its deposits are 235 and borrowing 25, leaving the same equity of 30. The receiving bank gains reserves and deposit liabilities of 35. The central bank's loan asset and newly created reserve liabilities are 25, not 35. Distinguish the creation of settlement balances from their subsequent redistribution. These are additional paper calculations, not an extra scenario in the ledger player. They assume the requested transfer is otherwise valid and do not test a real bank's access to funding.

Now isolate fiscal financing with another paper sequence. A treasury with zero opening cash sells a bond for 90 to a resident nonbank. The buyer's deposit falls by 90 and its bond asset rises by 90. Its bank transfers reserves of 90 to the treasury's central-bank account. The bank loses an asset and a deposit liability together. No household has yet received support. The financing step changes the form of private claims without by itself creating private net wealth. The treasury must have the legal spending and borrowing authority assumed for this sequence.

The treasury next pays support of 90 into eligible recipients' bank accounts. Its central-bank account falls by 90, recipient banks gain reserves of 90, and recipients gain deposits of 90. Across issuance and payment, private deposits and bank reserves return to their initial aggregate totals, while the private sector holds 90 more government bonds. Distribution has changed because the bond buyer and support recipients need not be the same people. With no taxes or other flows here, the public deficit is 90 and private net financial claims on the consolidated public sector rise by 90. This does not establish how much recipients consume.

Contrast a central bank buying an already outstanding bond from a resident nonbank for 90. The investor swaps a bond worth 90 for a deposit of 90. Its bank gains reserves and a matching deposit liability. There is no fiscal transfer of 90 to that seller at the transaction price. Buying the bond from a commercial bank's own portfolio instead replaces its bond asset with reserves without requiring a customer deposit increase. Those counterparty differences explain why two announcements with the same headline amount can have different deposit effects. They do not provide a calibrated estimate of either operation's effect on inflation.

Work through the accounts and assumptions

Observed eventAccounting conclusionAdditional question
Mortgage grantedLoan and deposit can rise togetherWhat is purchased and can supply respond?
Emergency reserve loanBank assets and debt both riseIs the bank solvent and eligible?
Bond issue then fiscal supportTrace both steps separatelyWho spends, on what, with what capacity?
Inflation slowsRate of price increase fallsDid the price level actually fall?
The case is invented. No row supplies a calibrated inflation forecast or a recommendation for a real country.

Use the evidence table to classify conclusions

The table above summarizes the invented case rather than displaying a numerical simulator. Read each accounting conclusion as a claim about specified entries, then answer the separate question in its final column. For the mortgage row, a larger loan asset and deposit liability follow from the stated origination. Whether the borrower buys an existing property or commissions construction remains open. For the emergency-loan row, reserve assets and borrowing rise together. Eligibility and solvency still need evidence. A balanced pair of entries establishes consistency, not an institution's legal permission to transact or its ability to survive further losses.

A useful audit distinguishes four kinds of statement. A definition specifies the object, such as annual consumer inflation. An accounting statement reconciles recorded amounts. An institutional statement describes a rule, such as which banks can obtain a facility loan. A behavioral statement predicts responses, such as borrowers reducing purchases after a rate increase. These statements can depend on one another, but they need different support. A ledger can verify the counterpart of a reserve loan. It cannot verify a predicted reduction in consumption. A law can establish authority without proving the size of a spending response.

Practice the distinction with an additional paper observation. A bank reports deposits of 800 at the start of a quarter and 880 at its end. The increase is 80, or 10% of the opening stock. That observation does not identify 80 of new mortgage lending. New bank asset purchases can create deposits, principal repayments can remove them, and payments involving other sectors can change the measured total. To reconcile the stock, imagine deposit-creating entries of 120 and deposit-reducing entries of 40, with no other changes. Their net is 80. Gross transactions and a net stock change answer different questions.

Apply the same reconciliation discipline to public debt. Suppose opening debt is 600, the overall deficit is 30, and the treasury increases its cash buffer by 10 using extra borrowing. With no other stock-flow adjustments, ending debt is 640. Inferring a deficit of 40 from the debt increase would mistake cash accumulation for expenditure. If ending annual nominal GDP is 800, the ending debt ratio is 640 divided by 800, or 80%. Dividing by an opening GDP of 750 instead gives about 85.33%, but that is a different comparison. Every ratio needs a dated numerator and a stated annual denominator.

For rate claims, record the contract and timing too. A policy rate increasing today does not immediately reset every old fixed-rate bond or mortgage. A bond price is today's valuation of payments at future dates, while a yield is the annualized return consistent with that price and those payments. If market-required yields rise, fixed promised payments generally command lower prices. The issuer need not have defaulted. For the invented bank, therefore, ask whether long bonds must be sold to meet withdrawals and how they are valued. A reserve loan may avoid a forced sale without reversing an already recognized asset loss.

Ask for evidence that can distinguish explanations

An expert claim should identify the money measure, sector, currency, period, and institutional regime. Reconcile stocks and flows before inferring causation. Look for credit volumes and uses, loan terms, bank capital and liquidity, import prices, inflation breadth, wages, output, expectations, and fiscal design. Distinguish an announcement from implementation and a forecast from an observed outcome.

Finally state what would change your conclusion. Falling energy inflation with a persistently high price level is different from restored cheap energy. Deposits rising after a central-bank asset purchase do not prove new bank loans caused them. A bank with liquidity support can still have insufficient capital. You should now be able to follow these distinctions in expert discussion while recognizing when institutional detail, econometrics, or specialist legal analysis is needed.

Specify observations that could reject a causal story

A causal claim compares what happened with what would have happened without a specified action. That alternative is the counterfactual, and it is not directly visible for the same country at the same time. Suppose annual inflation is 8% before a policy-rate increase and 5% afterward. One story says the rate increase reduced spending pressure. Another says the energy disruption ended while policy had little effect over that interval. Both can fit the two inflation numbers. To distinguish them, examine timing, sector prices, credit conditions, spending, and evidence about the shocks prompting the policy decision.

A comparison country helps only if the comparison addresses relevant differences. If one country depends heavily on imported fuel and another produces its own, different inflation paths may reflect fuel exposure rather than different monetary actions. If policy rises more where inflation pressure is strongest, a positive correlation between rates and inflation can reflect the central bank's reaction. A useful study must explain how it separates the action from those conditions. The lesson does not implement that statistical analysis. Its contribution is to identify what a proposed design must distinguish before attaching a causal interpretation to an observed association.

The distinction between rates and levels also changes what counts as contrary evidence. In a new paper example, a price index starts at 100, reaches 108 after one year, and 113.4 after the next. Inflation is first 8%, then 5%, because the second rise is 5.4 divided by 108. Prices have risen a cumulative 13.4% since the start. The second year shows disinflation, not deflation or restoration of the original basket cost. A claim that households should now face cheaper purchases does not follow from slower inflation. Income changes would need their own comparison to assess purchasing power.

State potential rejection criteria before selecting supporting facts. If someone claims reserve creation mechanically forces unlimited lending, evidence of banks holding more reserves while turning down risky borrowers challenges the proposed necessity. If someone claims the energy disruption explains every price increase, examine sectors with little direct energy use and the timing of their changes. Indirect energy inputs can still matter, so that observation alone is not decisive. The point is to name an exposure and trace a mechanism that could be checked. A theory adjusted after every observation without specifying any possible contradiction supplies little testable content.

Finish an audit with conclusions of different strength. The mortgage can create a loan and deposit together under the stated entries. The reserve loan supplies liquidity and matching debt, not equity. Fiscal financing and support must be traced separately, and their distribution matters. The size of any inflation effect remains conditional on spending, capacity, contracts, and other shocks. Keep the floating own-currency regime in view rather than importing the constraints of a foreign-currency debtor or a currency-union member. This written separation tells another reader exactly which results follow from accounting and where additional institutional or empirical evidence would change the assessment.

Exercises

Diagnose the bank, not the headline

A bank has assets 150, liabilities 140, and equity 10. Its assets suffer a fresh loss of 18, and it then borrows 25 reserves from an eligible facility in this hypothetical case. A headline says the extra reserves made it healthy. Audit the claim.

Hint

Calculate equity before and after the liquidity loan.

Worked answer
StageAssetsLiabilitiesEquity
Opening15014010
After loss132140−8
After reserve loan157165−8
Funding cannot undo a loss merely by expanding both sides of the balance sheet.

After the loss, assets are 132 and liabilities 140, so equity is negative 8. The loan raises assets to 157 and liabilities to 165, leaving equity negative 8. It can improve payment capacity but does not repair solvency. Actual eligibility for such support is not established by the numbers, and would require legal and supervisory assessment.

Audit a causal story

After a central bank buys bonds from resident nonbanks, deposits rise. Consumer inflation falls from 8% to 3%, and mortgage lending slows. Someone claims the extra reserves caused unlimited new mortgage lending and lower consumer prices. Identify what is supported and what is not.

Hint

Separate counterpart entries, quantities observed, and causal inference.

Worked answer
ClaimAssessment
Purchases can add depositsConsistent with nonbank sale entries
Reserve increase proves mortgage expansionFalse mechanism; observed lending slows
3% inflation means falling pricesFalse; prices still rise
Purchases caused the inflation changeNot identified from these observations
The valid accounting inference is narrower than the proposed macroeconomic story.

The purchase can directly create deposits and reserves without any new mortgage. Slowing mortgage lending contradicts the claim of unlimited realized lending. Inflation falling to 3% means prices still rise over the measured interval, not that they fell. None of these observations alone identifies the causal effect of the purchases; alternative shocks, policy expectations, and a counterfactual must be examined.

Sources and further reading

These references support the mechanisms and identify institutional context. They are optional; the explanations and worked answers above stand alone. Numerical examples are teaching cases, not live financial data or personal financial advice.

Lesson browserHow modern monetary systems work lessonsPrevious: Exchange rates and monetary constraints