What macro means for a trader

One number, published at a stated minute 

The number itself is a measurement of a period that is over. Prices collected in shops last month, hours worked last quarter, goods loaded onto ships and counted at a border. It contains no instruction, it commits nobody to anything, and the economy it describes carries on exactly as it was in the second after publication. What changes at that instant is not the world. It is what is known about the world.

Before the release, participants held a view about the condition being measured. After it, that view is updated, and every price set on the basis of the old view is set again on the basis of the new one. One sentence states the whole of why that happens, and it is worth having before any institution or release is named: an economic statistic reaches a price by way of a policy decision that has not been taken yet. The route is short, it is the same route every time, and each lesson in this module attaches to one link of it.

The chain 

The chain runs from a condition in the real economy, to a measurement of it, to the response of the institution charged with responding, to the return available on money held in a currency, to the price of that currency against another. Set out as steps, it looks like this.

1. A condition exists. Prices are rising faster or more slowly than before, employment is expanding or contracting, output is growing or shrinking. The condition is continuous and real, and nobody observes it directly.
2. A measurement of the condition is published on a schedule, by a statistics agency, a ministry or a central bank. It is partial, it describes a period already past, and it is frequently corrected later. It is also the only thing the market ever sees, which is why the gap between a measurement and the condition it measures has a lesson of its own further into this module.
3. A central bank whose mandate is written in terms of that condition reads the measurement alongside everything else in front of it, and either moves its policy rate, indicates that it may, or does neither. Doing neither is a decision and is priced as one.
4. The policy rate sets the return available on money held in that currency, and it anchors the structure of interest rates built on top of it, from overnight lending to long dated government borrowing.
5. Capital compares the return available in one currency against the return available in another, adjusted for what inflation is expected to do to each, and an exchange rate is where that comparison settles at any given moment.
6. The same rate is the rate at which a stream of company earnings expected in the future is converted into a value today. That is the second exit from the chain, and it is how a decision about a currency reaches the price of an index.

The conditions at step one do not drift independently of one another. Growth, employment and inflation move together in a long and irregular rotation, in which an expansion generates the price pressure that eventually invites a policy response, and the response works through the economy slowly enough to land in conditions different from the ones that prompted it. No two rotations share a length or a shape, which is why the word cycle describes a pattern rather than a schedule.

Two properties of the chain matter more than its steps. The first is that it runs on expectation rather than on events. A price already contains what participants collectively expect at every link, including the decision the central bank has not yet taken, so a release moves prices only to the extent that it differs from what was already assumed. A universally anticipated decision can pass with little movement, and a statistic nobody anticipated can move more than the decision it eventually contributes to. That relationship has its own lesson later in this module.

The second is that no link is mechanical. Step three is a judgement made by a committee of people who disagree with one another, published in language that is itself interpreted. Step five is a comparison made by participants with different horizons, obligations and reasons for holding a currency, some of which have nothing to do with the return on it. The chain describes where the pressure comes from, not an arithmetic that resolves to one number.

One piece of arithmetic in the chain is worth doing explicitly, because it is why a rising policy rate and a falling currency are not a contradiction. The return on holding money is conventionally compared after inflation rather than before it, and that comparison starts from the difference between a nominal rate and an inflation rate over the same period.

Both cases carry the same nominal rate. What separates them is the inflation figure beside it, which is why a policy rate quoted alone is an incomplete description of the return on a currency, and why an inflation release can matter to a currency price in a week when no committee meets.

Why this is not only a currency subject 

Foreign exchange is where the chain is most visible, because a currency pair is literally the ratio of two of them and a pair quotes the comparison directly. It is not where the chain stops. The sixth step, discounting, is what carries the same decision into every other class of contract, and it does so through one calculation that is small enough to write out.

The amount expected in the future did not change between the two cases. Only the rate used to convert it changed, and the value today changed with it. That is the mechanism by which an index can move on the publication of a labour market statistic while no company has reported anything at all, and it is how the chain reaches each of the underlying markets a contract can be written on, which are set out across the markets pages.

• Currency pairs price the comparison itself. A pair is two policy settings, two inflation conditions and two sets of flows expressed as a single number, which is why a release concerning one economy moves a price that names two.
• Index contracts track a benchmark whose value rests on company earnings expected in the future. The rate those expectations are discounted at sits at the end of the chain, so an index carries a rate sensitivity belonging to no individual company.
• Shares contracts inherit that sensitivity and add their own. A company borrows at rates anchored by the policy rate, and its customers respond to the conditions the statistics measure, so discounting and demand reach one price by two routes.
• Commodities and metals are quoted in a currency, so a move in that currency changes the quoted number without anything changing in the physical market. Rates additionally affect the cost of holding inventory, and the growth conditions the data measures affect how much of the commodity is consumed.
• Exchange traded fund contracts read the price of a listed fund, which aggregates holdings that each carry their own sensitivity, so the fund inherits a weighted version of what its holdings are exposed to rather than a sensitivity of its own.

So a release relevant to one economy is rarely relevant to only one position, and two positions that look unrelated can turn out to be two expressions of the same rate expectation. The portfolio level version of that observation closes this module.

Scheduled information and unscheduled information 

Information reaches a market in two ways, and what separates them is not importance. It is whether the arrival time was known in advance. A scheduled release has a publishing body, a date, a time to the minute and a stated coverage period, all announced months ahead and rarely moved. A committee decision has the same properties, and so does the press conference that conventionally follows it.

Because the timing is public, everything around the release adapts to it. Positions are adjusted beforehand by participants who would rather not hold them through it. Firms quoting prices widen or withdraw their quotes as the minute approaches, because quoting a two sided price into information nobody has yet is the one moment in the day when a market maker cannot be better informed than the participant on the other side. The most predictable thing about a scheduled release is therefore not the number. It is the state of the market around it.

Unscheduled information is everything else. A pipeline stops, a bank fails, a policymaker says something unplanned into a microphone, a government falls. No calendar contains it, no positioning precedes it, and it arrives into whatever depth the market happens to have at that hour, which the sessions module described as a function of who is at work. The distinguishing property is not severity. It is that the minute it arrives in is not distinguishable in advance from an ordinary one.

The economic calendar is the published list of the first kind only, and reading it begins with knowing what it is a list of. It is not a list of what will happen and not a list of what will move. It is a list of moments at which information will exist that does not exist now. How that list is laid out, and how the same release is described differently by different providers, is the subject of its own lesson further into this module.

Why this is a risk subject before it is anything else 

Two questions can be asked of a calendar entry, and only one of them has an answer. The answerable one is what the conditions of execution are likely to be around that minute: whether the cost of transacting widens, whether depth thins, whether a price can jump rather than travel. That is answered by the structure of the market itself, it is the same answer for every participant, and it is knowable beforehand because it does not depend on the release at all.

The unanswerable question is what the price will do. It is unanswerable for a structural reason rather than a modest one: the price already contains the consensus expectation, so what remains to move it is the part nobody anticipated, and the part nobody anticipated is by construction not knowable in advance. A schedule of releases is therefore a schedule of moments when the distribution of outcomes is wide. It is not a schedule of opportunities, and nothing in this module treats it as one.

That is the sense in which macro literacy is a risk skill first. It states when the plumbing of the market changes, and says nothing about which way anything is going. The lessons that follow on individual institutions and releases explain what each one measures and how it is conventionally interpreted, never what any of them implies for a position.

Where practitioners disagree 

Two arguments run underneath everything that follows, and a reader will meet both. The first is whether scheduled releases carry usable information at all. One tradition treats them as the primary events in a market's week, on the grounds that they are the moments at which genuinely new information about a whole economy becomes public. Another argues that the number reaches everybody in the same instant, that automated participants incorporate it faster than a person can read it, and that what is left afterwards is mostly wider costs and thinner depth. Both positions are held by experienced people looking at the same releases, which is why the argument persists rather than resolving.

The second is how much of the chain actually runs. One view treats monetary policy as the dominant influence on a currency and reads a pair primarily through the difference between two policy settings. Another holds that flows unconnected to rates govern for long stretches: trade balances, reserve management, sovereign investment, hedging by companies with foreign obligations, and the positioning of participants relative to one another. The second view has a particularly strong case where an exchange rate is administered rather than left to float, because there step five does not settle in a market at all and the pressure the chain describes emerges somewhere else. This module treats both views, gives the regional central banks the same standing as the larger ones, and returns to administered exchange rates in a lesson of their own.