International Macroeconomics
NX = −NKI. Every trade-deficit and tariff question goes through net capital inflows first. And appreciation is a rise in the price of the currency, so ¥160 → ¥150 per dollar is a weaker dollar.
Vocabulary and the market for dollars
- Exchange rate
- The price of one currency in another.
- Appreciation
- That price rises. A "strong" currency.
- Depreciation
- That price falls. A "weak" currency. ¥160 → ¥150 per dollar: a dollar buys fewer yen, so the dollar depreciated and the yen appreciated.
- Demand for dollars
- Foreigners buying American goods, services or assets.
- Supply of dollars
- Americans buying foreign goods, services or assets.
The balance of payments identity
(EX − IM) + (KI − KO) = 0
NX = −NKIKI capital inflows (foreigners buying US assets), KO outflows. Export on net and you are buying assets abroad; import on net and you are selling assets abroad.
NKI rises with the US real interest rate, falls with foreign real rates, and moves with tastes for US versus foreign assets. So a rise in r raises NKI, and since NX = −NKI, a rise in r lowers NX — completing Unit 20's claim that C, I and NX all fall when r rises.
The long run: Law of One Price and PPP
Short-run exchange rates are set by demand for currencies. In the long run a different discipline takes over for shippable goods — the Law of One Price: with minimal transport costs and no trade barriers, the same traded good costs the same everywhere once converted, or arbitrage erases the gap.
Purchasing power parity generalises this to a basket, linking two countries' inflation rates to their exchange rate: persistently higher inflation ⇒ a depreciating currency, roughly enough to keep the basket costing the same in both places — the same PPP used in Unit 17 for China. Two cautions: it is a long-run tendency, not day-to-day rates, which are driven by asset demand; and it covers traded goods only — haircuts and rent do not arbitrage, which is why PPP-adjusted GDP differs from market-rate GDP. Formal treatment: ECON 100B.
The long-run real interest rate
If investment exceeds saving, a country must import capital and run a trade deficit. The US has looked like this since the 1960s. China's surpluses reflect its very high domestic saving.
Three wrong explanations of the trade deficit
| Theory | What it actually predicts |
|---|---|
| US goods are low quality | A weak dollar |
| Americans have strong tastes for foreign goods | A weak dollar |
| Foreign countries are protectionist | A weak dollar |
| US assets look highly attractive — tech in the 1990s, housing in the early 2000s, perhaps AI now | NKI large and positive, so NX large and negative, with a strong dollar |
All three popular theories would raise the supply of dollars and cut demand, so the dollar depreciates — which pushes NX back toward balance. Only the last survives: foreigners buy more US assets (demand out), Americans buy fewer foreign ones (supply in), the dollar appreciates, NX falls. Strong dollar and trade deficit are the same phenomenon.
The tariff application
Long-lasting tariffs on many goods, no retaliation. Short run: tariffs cut US purchases of foreign goods, so the supply of dollars falls and the dollar appreciates. At a given exchange rate imports fall and NX rises; the appreciation pushes the other way. Resolve it with NX = −NKI: tariffs do not move net capital inflows, so nothing happens to NX — and the long-run S + NKI = I diagram agrees, leaving r* and NX* unchanged. All models are approximations, so the effect is not literally zero; but the appreciating dollar is a genuine offsetting force, so as long as NKI does not respond much to the exchange rate, the impact on NX is small.
Rules that earn points
- Decide who needs whose currency. Foreigners buying US ⇒ demand for dollars. Americans buying abroad ⇒ supply.
- Interest rate changes shift both curves the same direction on price. Say so.
- For any net exports question, go through NX = −NKI. Figure out NKI first.
- For long-run r*, draw S + NKI against I.
- Getting appreciation backwards. ¥160 → ¥150 per dollar is a weaker dollar.
- Explaining the trade deficit with tastes or quality. Those predict a weak dollar.
- Concluding tariffs shrink the trade deficit without checking NKI.
The yen-dollar rate falls from ¥160 to ¥150 per US dollar. Which is correct?
- AThe dollar appreciated and the yen depreciated
- BThe dollar depreciated and the yen appreciated
- CBoth currencies depreciated
- DNeither moved; only the quote changed
Answer
B — the dollar depreciated, the yen appreciated.
A dollar now buys only ¥150 instead of ¥160, so it is worth less in yen. Equivalently, a yen now costs more dollars.
A is the trap, and it catches people who see a falling number and reach for "falling currency". The quote is yen per dollar, so the number falling means the dollar is cheaper. Always read which currency is in the denominator before you answer.
The US real interest rate rises. In the market for dollars:
- ADemand shifts out and supply shifts in; the dollar appreciates
- BDemand shifts in and supply shifts out; the dollar depreciates
- COnly demand shifts out; the effect on the exchange rate is ambiguous
- DBoth curves shift out; the exchange rate is unchanged
Answer
A — demand out, supply in, dollar appreciates.
Foreign savers want US bonds and must buy dollars first (demand out). American savers now also prefer US bonds, so they convert fewer dollars into foreign currency (supply in). Both push the price of the dollar up, so the exchange rate effect is unambiguous even though the quantity traded is not. A stronger dollar then makes US goods dear abroad and foreign goods cheap at home, so NX falls.
C is the trap: it remembers only the foreign side. Missing the supply shift is what makes students think the answer is ambiguous.
A country runs a large, persistent trade deficit. It follows that:
- ANet capital inflows are large and positive
- BNet capital inflows are large and negative
- CIts currency must be weak
- DIts saving must exceed its investment
Answer
A — NKI is large and positive.
NX = −NKI is an accounting identity, not a theory. A trade deficit is financed by selling assets abroad, necessarily. Equivalently, from S* + NKI* = I*, investment exceeds saving — which rules out D.
C is the trap and it is the whole point of the unit: the popular tastes-and-quality stories predict a weak currency, but the US runs its deficit alongside a strong dollar. Strong dollar and trade deficit are the same phenomenon, both driven by attractive US assets.
The US imposes long-lasting tariffs on many goods and nobody retaliates. Net capital inflows are unaffected. Net exports:
- ARise substantially, since imports become expensive
- BFall, since the dollar appreciates
- CAre essentially unchanged
- DRise in the short run, then fall back in the long run
Answer
C — essentially unchanged.
NX = −NKI. If NKI does not move, NX cannot. The two direct effects cancel: at a given exchange rate imports fall and NX rises, but the tariff cuts the supply of dollars, the dollar appreciates, exports fall and imports rise. The long-run S + NKI = I diagram agrees — tariffs move none of S, I or NKI, so r* and NX* are unchanged.
A is the trap and it is the everyday intuition. It stops at the direct effect and never checks the exchange rate. Worth adding: all models are approximations, so the effect is unlikely to be literally zero — but the offsetting force is real, so the impact is small.
(a) Derive NX = −NKI from the market for dollars and state what it means in words. (b) Three popular explanations of the US trade deficit are that US goods are low quality, that Americans have strong tastes for foreign goods, and that foreign countries are protectionist. Show with one diagram why all three fail. (c) Give the explanation that survives, and confirm it with the saving–investment diagram.
Model answer
(a) At the equilibrium exchange rate, dollars demanded equal dollars supplied:
In words: if you export on net, you are accumulating claims on foreigners — buying assets abroad. If you import on net, you must be selling assets abroad to pay for it. A trade deficit is financed by a capital account surplus, necessarily and by accounting.
(b) All three predict a weak dollar, not a large trade deficit. If US goods were poor and Americans craved foreign ones, the supply of dollars would rise and demand for dollars would fall, so on the FX diagram the dollar depreciates — and a cheaper dollar pushes net exports back toward balance. They cannot explain a persistent deficit running alongside a strong dollar.
(c) Since NX* = −NKI*, a persistently large negative NX requires a persistently large positive NKI. US assets look highly attractive to global investors — tech in the 1990s, housing in the early 2000s, perhaps AI now. On the FX diagram, foreigners buy more US assets (demand out) and Americans buy fewer foreign assets (supply in), so the dollar appreciates, and the strong dollar is what pushes NX negative. From S* + NKI* = I*, the same thing said differently: US investment exceeds US saving, so the country must import capital and run a trade deficit. That has been true since the 1960s. China's surpluses are the mirror image, reflecting its very high domestic saving.
- Starting from dollars demanded = dollars supplied, not asserting the identity.
- The one-sentence translation: a trade deficit means selling assets abroad.
- Saying all three wrong theories predict a weak dollar, and showing both curves moving on the FX diagram.
- Noting that a weak dollar would correct the deficit — that is why the theories are self-defeating.
- Tying the surviving explanation to NKI, and confirming it with I > S.
(a) State the Law of One Price and compute what a computer priced at $1,000 in the US should cost in Japan when the exchange rate is ¥155.96 per $1. (b) What does purchasing power parity imply about a country with persistently high inflation? (c) Give the two cautions the course attaches to PPP.
Model answer
(a) With minimal transport costs and no trade barriers, the same traded good must cost the same everywhere once converted into a common currency — otherwise buying low and selling high would erase the gap.
(b) PPP generalises the Law of One Price from one good to a basket, implying a relationship between two countries' inflation rates and the exchange rate between them. A country with persistently higher inflation should see its currency depreciate, roughly enough to keep the basket costing the same in both places.
(c) First, PPP is a long-run tendency, not a description of day-to-day exchange rates, which are driven by demand for currencies to buy assets. Second, it applies to traded goods only — haircuts and rent do not arbitrage across borders, which is exactly why PPP-adjusted GDP differs from market-rate GDP, as in the Unit 17 China–US comparison. The formal treatment is ECON 100B.
- The arbitrage argument stated as the reason the law holds, not just the formula.
- The arithmetic shown: 1,000 × 155.96 = 155,960.
- High inflation ⇒ depreciation, with the basket-equalising logic.
- Both cautions: long run only, and traded goods only.
- Connecting the non-traded caution to why PPP-adjusted GDP differs from market-rate GDP.