Open Economy Guide to Trade and Exchange Rates

The open economy gives us a puzzle worth solving. In 2025, the United States ran a bigger deficit on goods than a year earlier. Yet its overall trade deficit narrowed. How can both statements be true? Services supply the missing piece. Their surplus grew by $27.6 billion, enough to offset a $25.5 billion widening of the goods deficit in the same official release. BEA · 2025 annual trade release

A similar puzzle comes up when a currency loses value and a vacation abroad gets more expensive anyway. In that case, prices supply the missing piece. This guide works through both puzzles, then asks what changes when a country can influence the interest rate it pays. Along the way, you can check every numerical example with an ordinary calculator.

What makes an economy open?

An open economy trades goods, services or financial assets with the rest of the world. Consider three purchases by a household in North Carolina. A Brazilian-made coffee grinder enters US goods imports. Booking a hotel stay in São Paulo adds to US travel-service imports. By contrast, buying a newly issued Brazilian corporate bond creates a financial claim on a foreign borrower. Only the first two purchases directly enter net exports.

Location alone can mislead. For example, a Brazilian visitor who pays for a hotel room in North Carolina buys a US service export even though the hotel never ships anything overseas. Therefore, the relevant question is who supplies the service and where the buyer normally resides. Nationality and the currency on the receipt can both distract from that question. BEA · international transactions and account definitions

Why imports appear with a minus sign in GDP

Here, Y means domestic output, C means household consumption, I means domestic capital formation, and G means government purchases. Exports are X and imports are M. Suppose a retailer sells an imported grinder for $100 after paying $70 to import it this period. Consumption includes $100. Meanwhile, the import entry subtracts $70. Under these assumptions, the sale adds $30 to domestic GDP through the retail margin.

That subtraction prevents the accounts from counting foreign production as domestic output. Consequently, a larger import bill alone tells us little about the growth of domestic production. An expanding domestic industry, for example, may import more parts while increasing its own value added.

How saving and investment connect to trade

Start with a country that produces 1,000 units and consumes 650. Its government buys another 150, leaving 200 units of domestic saving. If firms invest 240 units at home, domestic spending exceeds production by 40. Therefore, net exports equal −40. As a result, the country can use more resources than it produces during this period because it can obtain financing from abroad.

In this simplified open economy, we set net international income and current transfers to zero. Under that assumption, national saving and domestic saving coincide. Net capital outflow (NCO) equals residents’ net acquisition of foreign assets less foreigners’ net acquisition of domestic assets. Thus a positive NCO corresponds to net lending abroad. A negative value corresponds to net financing from abroad, which can include equity as well as debt.

How much did households save, and how much did the government save?

Let T stand for taxes net of transfers. With net taxes of 180, private saving in the example equals 170. Public saving equals 30. Together, the two add up to the original 200. Now suppose taxes fall by 20 while purchases and output stay fixed. If households save the entire tax cut, private saving rises by 20 and public saving falls by 20. National saving holds steady.

Alternatively, suppose households spend 15 of the tax cut. Private saving rises by just 5. As a result, national saving falls by 15. A government budget deficit and an external deficit can move together, but the household response matters. The accounting identity can check either outcome. It cannot tell us how households will respond.

Where the current account enters the calculation

Primary income includes returns such as dividends or interest received from abroad, less corresponding payments. Secondary income includes current transfers such as personal remittances. For example, a worker who sends money to a parent living overseas makes a transfer. A visitor who pays that parent to operate a tour buys a service. The cash may cross the same border, yet it enters a different account.

For empirical work, national saving uses gross national disposable income. With that definition, national saving minus domestic investment corresponds to the current account. To reconcile the current account with measured net financial flows, also account for the capital account and the statistical discrepancy. For example, the official “capital account” records items such as capital transfers. Most international purchases of bonds and shares enter the financial account. BEA · international transactions and account definitions

A US trade deficit you can reconstruct two ways

The table compares two dated snapshots. Each agency release can revise earlier estimates as new information arrives. Read down either numerical column to reconstruct a balance. For a year-over-year change, use the comparison that the agency reports within a single release. BEA · 2024 annual trade release BEA · 2025 annual trade release

To check 2025, subtract imports from exports. The result is 3,432.3 − 4,333.8 = −901.5 billion dollars. Next, add the balances by sector. The result is −1,240.9 + 339.5 = −901.4 billion. Why does the second method miss by $0.1 billion? Each published component has only one decimal place. The two routes agree within the precision of those inputs.

A revision large enough to change a headline

Comparing the two original annual releases suggests the deficit narrowed by $16.9 billion from 2024 to 2025. However, the 2025 release gives a revised 2024 deficit of $903.5 billion and reports a narrowing of just $2.1 billion using underlying precision. Consequently, the revision accounts for most of the apparent improvement. BEA · 2024 annual trade release BEA · 2025 annual trade release

It also affects the measured growth of the services surplus. Using the two original releases, the services surplus appears to grow by $46.2 billion. Within the later release, however, the surplus grows by $27.6 billion. Both calculations can be reproduced, but only the second holds the release date constant. BEA · 2024 annual trade release BEA · 2025 annual trade release

Why a trade table cannot tell you the whole financing story

In its March 2025 release, the Bureau of Economic Analysis put the 2024 US current-account deficit at about $1.13 trillion. Secondary income receipts totaled $201.7 billion, while payments reached $408.6 billion. Consequently, current transfers alone contributed roughly −$206.9 billion to the balance. That helps explain why the current-account shortfall exceeded the trade shortfall. BEA · international transactions and account definitions

A deficit is a flow over a period. By contrast, an international investment position measures outstanding assets and liabilities on a date. Exchange-rate movements and changes in asset prices can alter that position even if nobody buys a new asset. Therefore, comparing two balance-sheet totals will generally give a different result from adding up a year’s trade transactions.

What a nominal exchange rate actually measures

Suppose a dealer quotes 5 Brazilian reais per US dollar. A $100 bill then buys R$500 before fees. Now let the quote rise to 6 reais per dollar. Each dollar buys more reais, so the dollar appreciates against the real. From a Brazilian buyer’s perspective, a dollar now costs more.

The same move can produce two different percentages. In reais per dollar, the quote rises by 6/5 − 1 = 20%. Conversely, the quote in dollars per real falls from $0.20 to about $0.1667, a decline of 16.67%. Both statements describe the very same transaction. Inverting a quote changes the denominator of its percentage change.

How to calculate a real exchange rate

Take the United States as home. Let e mean foreign currency per home dollar, P the home basket price in dollars, and P* the foreign basket price in foreign currency. Then ε measures the foreign baskets that have the same value as one home basket. A higher ε means a real appreciation of the home currency in this article. IMF · real exchange rates

For example, a US basket costs $100 and its Brazilian counterpart costs R$400. At 5 reais per dollar, the US basket costs R$500. Consequently, its relative price equals 500/400 = 1.25 Brazilian baskets. If the Brazilian basket rises to R$500 while both other inputs stay fixed, the relative price falls to 1.00. The US dollar has depreciated in real terms even though the dealer’s quote never moved.

When both P values are price indexes, their base periods matter. Two national CPI indexes of 100 tell us only that each series equals its own base value. Therefore, plugging those two numbers directly into a level comparison cannot tell us how many reais a US basket should cost. Matched basket prices support level comparisons. Consistently constructed indexes support comparisons of changes over time. IMF · real exchange rates World Bank · size of economies

Bilateral rates and trade-weighted indexes answer different questions

A Brazilian exporter selling only to the United States may care most about the bilateral real-dollar rate. By contrast, a central bank studying competitiveness across many destinations may use a real effective exchange rate. The BIS combines bilateral rates using trade weights and adjusts for relative consumer prices in its real indexes. An increase in the BIS index indicates appreciation. For example, a reading of 110 against a base of 100 means a 10% increase since the base period. Deciding whether the currency is expensive requires an additional benchmark. BIS · effective exchange rates

Purchasing power parity, inflation and a vacation in Mexico

Absolute purchasing power parity (PPP) says that equivalent baskets cost the same after currency conversion. Under our convention, the corresponding level is e = P*/P and ε = 1. Relative PPP asks a narrower question. If the relative basket price stays constant, how must the nominal rate change as the two national price levels grow?

Here π is home inflation and π* is foreign inflation over the same interval. Under relative PPP, ε₁/ε₀ equals one even if ε itself stays at 1.25. For example, foreign inflation of 10% and home inflation of 2% imply a rise in foreign currency per dollar of 1.10/1.02 − 1 = 7.84%. Subtracting the inflation rates gives an approximation of 8%. The gap grows when inflation is high.

Can the dollar buy more pesos but fewer meals?

Imagine a Mexican meal costs 100 pesos when one dollar buys 10 pesos. Its dollar price is $10. Over the next decade, Mexican prices double, US prices rise 25%, and the dollar reaches 15 pesos. These are classroom assumptions drawn from the supplied exercise. They do not describe a current travel quote.

First, divide 200 pesos by 15 to find the visitor’s dollar bill. Next, divide that bill by 1.25 to express it in start-year US purchasing power. Despite receiving 50% more pesos per dollar, the visitor pays about 6.67% more relative to US prices. For an unchanged relative price, the dollar would have needed to reach 10 × 2/1.25 = 16 pesos. That one-peso gap explains the entire real cost increase.

The Big Mac test in Mexico and Brazil

A single standardized product makes PPP easy to audit. In the historical 2005 exercise, the US reference burger cost $3.06. Mexico’s price was 28.1 pesos and Brazil’s was 5.95 reais. Divide each local price by $3.06 to obtain the exchange rate that equalizes burger prices. Then compare that implied quote with the observed local-currency-per-dollar rate. The following calculations use the rounded entries in the supplied table (page 144). The Economist · Big Mac data and methodology

For Brazil, the calculation is (5.95/2.49)/3.06 − 1. It measures the percentage difference between two burger prices after conversion. A burger incorporates local rent, wages, taxes and ingredients. Consequently, an international gap can persist even when restaurants use similar recipes. Shipping cooked meals across borders offers little opportunity to close it.

One more detail matters if you extend the calculation across decades. In July 2022, The Economist changed its US reference price from an average across four cities to a price supplied by McDonald’s. Therefore, anyone comparing the historical results with later observations should record the methodology and dataset version. The publisher provides both raw and GDP-adjusted measures. The Economist · Big Mac data and methodology

Why the World Bank counts more than hamburgers

To compare whole economies, the International Comparison Program uses comparable goods, services and expenditure data to measure purchasing power across economies. Its 2021 benchmark covered 176 participating economies at the GDP level. Together, their output reached about 152.4 trillion international dollars. China accounted for 18.9% and the United States for 15.5%. World Bank · size of economies World Bank · 2021 benchmark highlights (PDF)

For example, the Brazilian figure compares the volume of output after allowing for price differences. It does not tell a Brazilian importer how many US dollars a bank will deliver for a real. For a household consumption question, select consumption PPPs. To compare output, use GDP PPPs. In fact, the program publishes measures for 45 expenditure headings, so the choice can match the question. World Bank · PPP data by expenditure

Solving a small open economy step by step

A small open economy with perfect capital mobility takes the world real interest rate as given. In this long-run exercise, hold output fixed and let prices adjust. Firms invest until their desired investment equals I(r*), where r* is the world real rate. Consequently, the difference between saving and investment determines net exports. The real exchange rate adjusts until buyers want exactly that amount of net exports.

Use the following classroom schedule, adapted from the photographed exercise. All quantities use the same output units per period. Enter r as a number of percentage points, so a 5% rate means r = 5. The relative-price variable ε has no currency units.

Baseline, fiscal expansion and a higher world rate

First, calculate consumption. Disposable income equals 4,000, so households consume 250 + 0.75 × 4,000 = 3,250. At G = 1,000, national saving equals 750. With r* = 5, firms also invest 750. Therefore, NX equals zero and the exchange-rate equation gives ε = 1.

In the second row, government purchases absorb an additional 250 units. Since the world rate fixes investment, the saving shortfall shows up entirely in net exports. Domestic goods become relatively expensive until buyers demand 250 more units of imports than exports. In the third row, reset government purchases before changing the interest rate. Now domestic investment falls by 250, leaving the economy with a surplus of the same size.

What changes when households prefer imports?

Suppose buyers develop a preference for foreign cars. At every real exchange rate, desired net exports fall by 100. The new equation is NX = 400 − 500ε. However, baseline saving and investment still each equal 750. For balanced trade, ε must fall from 1.00 to 0.80. The relative-price adjustment restores the aggregate balance even though car imports can take a larger share of total spending.

Conversely, consider a quota that raises desired net exports by 100 at every ε. The revised equation is NX = 600 − 500ε. At zero aggregate net exports, ε rises to 1.20. As a result, exporters and firms competing with imports outside the protected sector face more expensive domestic goods. This outcome requires fixed saving and investment. A policy that also changes fiscal revenue, productivity or investment can produce a different balance.

How a large open economy shares a saving shock

Unlike a price-taking borrower, a large open economy can influence the interest rate at which it borrows. In the model, both domestic investment and net capital outflow fall as the domestic real rate rises. The first response reflects financing costs. Meanwhile, the second reflects the relative appeal of home assets to investors. The interest rate balances the uses of saving, and the real exchange rate balances net exports with net capital outflow.

To see how much adjustment falls on each market, give the economy 300 units of saving. Let I = 300 − 20r, NCO = 150 − 10r, and NX = 150 − 100ε. Equilibrium requires 300 = 450 − 30r. Therefore, r equals 5, investment equals 200 and NCO equals 100. The exchange rate that supports 100 units of net exports is ε = 0.50.

An extra 60 units of public spending

Now suppose government purchases rise by 60 while taxes, consumption and output stay fixed. National saving falls to 240. Solving 240 = 450 − 30r yields r = 7. Consequently, domestic investment falls from 200 to 160 and net capital outflow falls from 100 to 80. Net exports must equal 80, so ε rises to 0.70.

Why does investment absorb 40 units while the external balance absorbs 20? Each percentage-point increase in r cuts I by 20 and NCO by 10. Thus investment absorbs 20/(20 + 10), or two-thirds, of the lost saving. Net capital outflow absorbs the other third. Different slopes would split the very same fiscal shock differently.

Four policy experiments with the same starting point

In the next table, change one schedule at a time from the large-economy baseline. All four experiments use S = 300 unless the row changes saving. In particular, a safe-haven inflow lowers the NCO schedule because foreigners want more domestic assets at each interest rate.

Notice that the first two shocks below baseline generate the same interest rate, exchange rate and trade balance. Yet investment ends at 160 in one case and 220 in the other. Consequently, even perfect observations of exchange rates and trade balances cannot tell these shocks apart. Domestic investment provides a way to discriminate between the two explanations in this model.

The safe-haven example contains a second surprise. Initially, the shift reduces NCO by 60 at the old rate. However, the domestic rate then falls by two points, restoring 20 units of outflow along the new schedule. Net outflow ultimately drops by 40. This distinction between a shift of a curve and a movement along it matters whenever the interest rate can adjust.

What did the global saving glut hypothesis propose?

In a March 2005 speech, Ben Bernanke linked a greater global supply of saving to relatively low long-term real interest rates and a larger US current-account deficit. He presented a historical interpretation. It offers a useful comparison with the safe-haven example because both involve foreign funds becoming more available to domestic borrowers. Their underlying causes can differ. Federal Reserve · Bernanke’s 2005 speech

For example, stronger demand for the safety of one country’s assets can redirect an existing pool of wealth. Higher global saving can enlarge the pool itself. Looking only at one country’s trade deficit leaves both mechanisms plausible. Evidence about world interest rates, saving across economies and the composition of asset purchases can help distinguish them.

Three checks before you draw a conclusion

Could a high foreign interest rate guarantee a higher return?

Suppose a US saver can earn 8% on a dollar deposit or 12% on a Canadian-dollar deposit. Start with US$100 and assume one Canadian dollar initially costs one US dollar. The domestic account ends at US$108. Meanwhile, the foreign account ends at C$112. To match the domestic payoff, the final exchange rate must equal 108/112, or about US$0.9643 per Canadian dollar.

Therefore, a 3.57% fall in the Canadian dollar can erase the entire interest advantage. This is a break-even calculation. Actual returns depend on the realized exchange rate and any costs. Under equal expected real rates, the Fisher relation links higher nominal rates to higher expected inflation. Relative PPP then offers a reason to expect currency depreciation over time, although it gives no certain exchange rate for the withdrawal date.

Does a cheaper currency immediately improve trade?

Consider an importer that owes a foreign supplier €1 million under a fixed contract. At $1.10 per euro, the bill equals $1.10 million. If the dollar weakens to $1.20 per euro before settlement, the bill rises to $1.20 million even if shipment volume stays fixed. Over time, buyers may switch suppliers and exporters may win orders. Consequently, a long-run net-export schedule and a next-month invoice can move in different directions.

How would you check an inflation and exchange-rate claim?

Start by choosing one quotation, one time interval and matching price series. For example, use annual-average rates with annual-average prices, or year-end rates with corresponding end-period indexes. Next, compute the exact real change as (e₁/e₀) × (P₁/P₀)/(P*₁/P*₀) − 1. If the result is close to zero, relative PPP describes that interval reasonably well. One interval alone provides little evidence about the general theory.

For a longer study, calculate log changes for each interval. A plot of the change in the log nominal rate against foreign minus home log inflation shows the relative-PPP prediction as a line with slope one and intercept zero. Moreover, persistent deviations can point to changing relative prices, the choice of price index or timing differences. Recording both currency units and data vintages lets another reader reproduce the test. IMF · real exchange rates BIS · effective exchange rates

What to take from the open economy model

Return to the US trade puzzle. A widening goods deficit can coexist with a narrowing total deficit because services enter the same balance. Similarly, a dollar that buys more pesos can buy fewer meals because menu prices enter the real exchange rate. The arithmetic reveals which additional observation each headline needs.

When you encounter a new open economy claim, write down its units and reconstruct one transaction. Then identify the assumption that lets you move from an accounting result to a prediction. In the fiscal example, the same loss of saving generated three different investment outcomes. Knowing who sets the interest rate made it possible to tell them apart.