Capstone: Macroeconomics in Practice

This is Unit Eighteen of Principles of Macroeconomics, part three of three, and the last lecture in the course. Part one gave you seven models and the rule for choosing among them. Part two gave you the distinctions that run through all of them. This part does four things at once: it assembles the whole course onto a single diagram, maps the questions the exam publishes, sets out what the profession has not settled, and says what this course does not know. Carry one idea through. A course should end by saying what it does not know, and this one has a specific list — terms it taught from definitions because neither textbook named them, figures it refused to state because it could not stand behind them, and places where it corrected its own source from outside. Naming a gap is more useful than filling it badly, and that principle is what the closing entry is built on. By the end you will assemble the models onto one diagram, say why that diagram gives directions and never magnitudes, say what the published samples can and cannot tell you, sort a disagreement into three kinds, and name what this course leaves open.

One entry assembling almost everything the course taught onto a single diagram.

The whole course, assembled.

MAC 1737 The Whole Economy in One Diagram

Almost everything this course taught can be assembled onto the aggregate demand and supply diagram, and doing that assembly once is the most useful revision available. The horizontal axis is real output and the vertical is the price level. Long-run aggregate supply stands vertical at potential output, which unit four said is set by physical capital, human capital and technology, and which grows over time as those grow. Short-run aggregate supply slopes up, because wages and some prices are sticky, and it shifts when input prices, productivity or expected prices change. Aggregate demand slopes down and is the sum of consumption, investment, government purchases and net exports — so every one of the course's other models feeds one of its components. The expenditure model of unit nine determines the position of aggregate demand at a given price level. The money market of unit twelve sets the interest rate that moves investment. Loanable funds sets the real rate that government borrowing competes in. The foreign exchange market of unit thirteen sets the rate that moves net exports. They are not separate subjects; they are the machinery behind one curve. Trace one event through every model at once. A central bank raises the rate it pays on reserves. In the money market, banks hold reserves rather than lend, the federal funds rate rises, and other rates follow. In loanable funds, the real rate rises and some investment projects stop clearing their hurdle, so investment falls. In foreign exchange, higher domestic returns attract foreign funds, so demand for the currency rises and its supply falls, and it appreciates. In the goods market, the appreciation makes exports dearer abroad and imports cheaper at home, so net exports fall. Two components of aggregate demand have now fallen, so aggregate demand shifts left by the multiplier times their combined fall. On the diagram, output falls and the price level falls, and how much of each depends on which zone of short-run supply the economy was sitting in. Three cautions. First, the diagram gives directions and not magnitudes: it can say the price level falls without saying by how much, and every quantitative answer in this course came from a model underneath it rather than from the diagram itself. Second, when both curves move, the outcome for one variable is determinate and the other is not, and the honest answer to the indeterminate one is that it cannot be determined from the information given. Third, the long-run curve is doing quiet work throughout: everything that looks like a permanent gain in output is either a movement of that vertical line or a temporary deviation from it, and unit ten's whole argument is about which one you are looking at.

One entry on the questions the exam's publisher releases, and on what they can and cannot tell you.

Using the samples as a map.

MAC 1738 The Twelve Published Questions, Mapped

The exam's publisher releases twelve sample questions, and they are worth using as a map rather than as practice, because twelve is too few to practise on and exactly enough to show the shape of the thing. Their text is not reproduced here. What follows is what each one tests and where this course covers it. Two are on comparative advantage and trade, from a table of production possibilities and from a currency movement raising exports — units one, thirteen and seventeen. Two are on a single market, one where an expected future price shifts both demand and supply and one where a price floor creates a surplus — unit two. Two are on aggregate supply and the Phillips curve, one asking what shifts short-run supply right and one asking about movement along the short-run curve — units eight, nine and twelve. One is on a price index built from a fixed basket — unit six. One is on the spending multiplier — unit nine. One is on loanable funds, where more private saving lowers the real rate — units two and fifteen. One is on the central bank fighting inflation — unit twelve. One is on human capital and long-run growth — unit four. One is on interest-rate differences driving capital flows — units seven and thirteen. Look at what the distribution tells you. Of the twelve, four are single-market or frontier questions from the course's first two units, four are about aggregate demand and supply or policy, two are measurement, and two are open economy. None requires more than arithmetic. None asks for a definition in isolation; every one puts a term inside a situation and asks what follows. And at least three of them can be answered by part one's four-step method or its shock trace without any specialised knowledge at all. That is the strongest single piece of evidence for how this unit is built: the procedures generalise, and the published questions are drawn from the same small set of procedures the course has been drilling. Three cautions about using them. First, twelve questions cannot tell you the weights. The published outline does that, and the outline says stabilisation policy is the largest category while only two of the twelve touch it, so do not infer emphasis from the sample. Second, one of the twelve contains an option naming a policy tool whose value has been zero since March two thousand and twenty, which unit twelve stated plainly — the option is still wrong for the reason the question intends, and a student who knows the current fact reaches the same answer by a shorter route. Third, sample questions are illustrations chosen to be clear, and the real exam contains items designed to separate students, so the genuine article will be less tidy than these are.

One entry assembling every disagreement the course recorded, and sorting them into three kinds.

Three kinds of disagreement.

MAC 1739 Where Economists Disagree, and Why

This course recorded disagreement wherever it found it, and assembling those disagreements shows something about the subject that no individual unit could. They are not all the same kind. Some are empirical and unresolved: how large the fiscal multiplier is, with careful estimates in the same year running from near zero to around two; how much government borrowing crowds out private investment, where the studies are inconclusive and the answer appears to depend on the type of spending; and how far households offset public borrowing by saving, where the measured offset is about a third rather than the whole. Some are about horizon rather than fact: how fast an economy self-corrects, where the working estimate of two to five years is itself disputed, and where the Keynesian and neoclassical accounts are both internally consistent and differ mainly on how long the middle of the story takes. And some are about values wearing an empirical costume: whether to close a gap by spending or by taxing, which the textbook itself says is a political choice, and what labour standards should be enforced across borders. Take one disagreement apart. Ask whether expansionary fiscal policy works. A first economist says the multiplier is around one and a half, so it works. A second says it is near zero, because households save the transfer and interest rates rise. A third says the honest answer is that the available data cannot separate the two. All three are looking at the same episodes. They differ on what to control for, on whether the economy was near potential, on what the central bank was doing at the time, and on how much of the measured change would have happened anyway. Notice that the disagreement is not about the model — all three accept that aggregate demand shifts right — but about a magnitude the model does not supply. Almost every serious macroeconomic dispute has that shape. Three cautions. First, an exam question asks what a model implies, not which economist is right, and the model implies a direction; where a question turns on a magnitude that is genuinely disputed, it will give you the magnitude. Second, the existence of disagreement is not evidence that nothing is known: the direction of most of these effects is not in dispute, only the size, and a subject that can bound a question to its magnitude has made real progress. Third, be suspicious of confident answers to the value questions, including confident answers you find congenial — this course reported both sides of every one of them and settled none, and that was a decision rather than an omission.

One entry to close the course, on what it does not know and why saying so matters.

What the course leaves open.

MAC 1740 What This Course Did Not Settle

Here the unit's governing idea pays off. A course should end by saying what it does not know, and this one has a specific list. Three terms on the exam's own outline are taught here from their definitions, because neither of the course's two textbooks names them: the policy mix, which both books describe as a mechanism without ever calling it that; the balanced-budget multiplier, which the course derives from the expenditure line rather than quoting; and the present-value half of the time value of money, which neither book teaches. Several figures could not be verified and are therefore absent rather than approximated: the daily volume of the foreign exchange market, where the two books give three incompatible numbers; the World Bank's current income thresholds, where the textbook gives four inconsistent sets in one chapter; and the World Trade Organization's membership, which its own pages would not yield. Where a source was wrong the course said so and corrected it from outside — the reserve requirement set to zero in twenty twenty, the euro's eleven founding members, a trade agreement's date. And one unit, on trade and protectionism, rests on a single textbook across most of its range, which every entry there states in its own warning. Consider why that matters for using the course. Suppose you meet a question about the balanced-budget multiplier. This course derived it rather than quoting it, and said it equals one only in the simplest case — with no income tax and no imports — and showed that in a taxed economy the net effect is smaller. A course that had merely asserted that the balanced-budget multiplier is one would have left you confident and wrong in the more realistic case. The same applies to the income thresholds: a course that had quoted one of the textbook's four inconsistent sets would have given you a number to remember and no way to know it was contradicted twenty pages later. Naming the gap is more useful than filling it badly. Three closing cautions. First, an absent figure is not an absent concept: the course teaches what the foreign exchange market is, what the income tiers are, and what the World Trade Organization does, and declines only to state numbers it could not stand behind. Where you need a current figure, the course names the body that publishes it. Second, the textbooks behind this course are dated in specific ways that were recorded as they were found, and a fact that turned on a date was either sourced elsewhere or dropped. Third, and most importantly, nothing in this course tells you what an economy should do. It tells you what follows from what, which is the part that can be examined, and leaves the rest to you.

Look at what you can now do. You can assemble the whole course onto one diagram, and say that the other six models are not separate subjects but the machinery behind one curve. You can say why that diagram gives directions and never magnitudes, and why the honest answer to an indeterminate variable is that it cannot be determined. You can use the published questions as a map rather than as practice, and say why twelve of them cannot tell you what the exam emphasises. You can sort a disagreement into empirical, horizon, or values — and say that a subject arguing about the size of an effect rather than its direction has made real progress. And you can name what this course leaves open: the terms it taught from definitions, the figures it would not state, and the places it corrected its own source. That is the end of Principles of Macroeconomics. Nothing here tells you what an economy should do; it tells you what follows from what, and leaves the rest to you.