The economic cycle – where we are and how to measure it

The economic cycle is not textbook theory, it is a time frame that explains why the same strategy sometimes delivers 20% in a quarter and sometimes 8 losses in a row. The economy moves through phases: expansion, overheating, slowdown, recession, rebound. Every instrument, every sector and every strategy has its place in the cycle. If you do not know where you are, you do not know what to expect from your edge.

The goal of this lesson is to teach you to place today on the map of the cycle using measurable indicators, not the media narrative.

The 4 phases of the cycle – an operational definition for a trader

The traditional division is too slow for the market. The market discounts the phase 6-9 months in advance. We will use a division that combines the economy and the market reaction.

Important: the stock market most often rises in phases 1 and 2, peaks somewhere in 3, falls hardest in the 3->4 transition and bottoms at the start of 1. You do not want to guess the peak – you want to know which indicators confirm the transition.

The indicator set – your cycle cockpit

You cannot determine the cycle with one indicator. You need a cockpit with leading, coincident and lagging ones. The same thing investment banks do.

Leading – they look 6-12 months ahead:

• The 10Y-2Y and 10Y-3M spread – an inversion is an alarm, not a verdict.
• PMI New Orders - Inventories – a difference >0 = inventory restocking = growth.
• Building permits – a slowdown here is a slowdown in investment.
• S&P 500 vs the 200-day moving average – the market itself is a leading indicator.

Coincident:

• GDP, industrial production, retail sales, payrolls.

Lagging:

• Unemployment rate, CPI, overdue debt, the default rate.

For a trader the most important thing is the change in momentum, not the level. An economy with GDP at 1.5% can be in recovery (if a quarter ago it was 0.2%) or in slowdown (if a quarter ago it was 3%).

The formula for cycle momentum:

Cycle_Momentum = (PMI_t - PMI_{t-3}) + (NFP_3m_avg_t - NFP_3m_avg_{t-3})/100 + (Spread_10Y2Y_t - Spread_10Y2Y_{t-3})*10

Momentum >0 and PMI>50 -> expansion
Momentum <0 and PMI<50 for two months -> slowdown
Momentum strongly negative and Spread rising (steepening after an inversion) -> late slowdown / close to a rebound

This is not a ready forecasting model – it is a frame so that you ask yourself the same question every day and compare the answers.

Where we are – a measurement checklist

Instead of reading headlines asking “will there be a recession”, answer 7 questions. Each one yes/no, with a number.

Operational interpretation:

• 0-2 yes: expansion / early overheating
• 3-4 yes: late overheating / early slowdown – rotate into defensives, reduce beta
• 5-7 yes: slowdown / recession – capital is protected by long bonds and cash

An important rule: The cycle is not for timing to the day, it is for setting a risk filter. With 5-7 yes you do not increase directional long positions in growth stocks hoping for a miracle. You reduce your unit of risk.

Mapping sectors to the cycle

This is the direct usefulness of the cycle for an ETF trader.

Phase 1 Recovery: QQQ, IWM (Russell 2000), XLI (industrials), XLF (banks)
Phase 2 Expansion: XLE (energy), XLB (materials), XLI, EEM (emerging)
Phase 3 Overheat: XLE, XLU (utilities) starts to outperform, XLP (staples)
Phase 4 Slowdown: XLV (healthcare), XLP, XLU, TLT (long bonds), GLD (gold as a hedge)

Note: this is not “buy this” advice. It is a map of historical relationships. Your job is to check whether the rotation looks similar this time, and whether fiscal/monetary policy is distorting the cycle (see lesson 5).

When the plan says DO NOT TRADE

1. When the cockpit scores 3-4 yes – that is, a transition phase. This is the hardest moment. Mixed indicators, the market in a range, false breakouts. If your strategy needs a trend, there is no trend in a transition phase – there is rotation.
2. When “everyone knows where we are” – the narrative is unanimous. If 90% of commentators say “recession in 2 months” and positioning is already extremely defensive, the risk is the opposite – the absence of a recession will be the shock. Do not trade the consensus without measuring positioning (COT, AAII surveys).
3. When the global and the local cycle diverge and you trade an instrument that depends on both. For example, Poland in recovery but Germany/the US in slowdown – the WIG (the Warsaw Stock Exchange index) may rise, but export companies will not. Without separating the exposures it is a lottery.
4. When you try to catch the recession bottom with a single order. Bottoms are a process, not a point. In 2008, 2020 and 2022 the bottom was built over weeks. The plan should talk about scaling, testing and a stop, not about “buy the panic”.

The mistakes that cost the most

1. Extrapolating the phase. A trader made money in phase 2 on cyclical stocks and assumes it will always work from now on. In phase 4 the same stocks lose 40%. The cycle ends every edge that is not adaptive.

2. Confusing lagging with leading. “Unemployment is low, so there is no recession”. Unemployment is lowest right before a recession. It is a lagging indicator. Watching it is driving while looking in the rear-view mirror.

3. Not measuring your own results by cycle. You do not keep a journal tagged with the phase of the cycle in which you took the trade. After a year you do not know that 70% of your losses are phase 3-4 under a trend strategy that works only in 1-2. That is a cost a filter could have cut.

4. Linear thinking about recession. A recession is not 2 quarters of falling GDP by the NBER definition. It is a process of declining real activity, income and employment. Waiting for the “official recession” makes you 9 months late to the market reaction.

A measurable exercise [30 minutes]

Goal: Build your own cycle cockpit and tag today.

Steps:

1. Go to FRED and ISM, download the last 6 months: the 10Y-2Y spread, PMI Manufacturing, New Orders, Initial Claims, the NFP 3-month average, the High Yield Spread (BAMLH0A0HYM2).
2. Fill in the table from the 7-question checklist (yes/no). Count the total of YES.
3. Take your trade journal from the last 3 months – tag every trade with the current YES total from the day of entry (if you do not have a journal, tag 5 trades you remember).
4. Calculate:

WinRate_Phase_0_2 = Profitable / All when the total is 0-2
WinRate_Phase_3_4 = Profitable / All when the total is 3-4
WinRate_Phase_5_7 = Profitable / All when the total is 5-7

5. If the WinRate drops by >15pp in phase 3-4 versus 0-2, you have your filter: cut risk by 50% in phase 3-4 or trade only a defensive basket.

Pass criterion: You have a cockpit table for the current month and one number: the current YES total (for example “Today 4/7 – late overheating / early slowdown”) plus a conclusion for position size.

# Journal template
Date: 2024-12-13
YES_Total: 4/7
Phase: Overheat -> Slowdown
Plan: I cut the size of directional positions by 40%, only defensive longs / short high beta with confirmation allowed

Test yourself

1. The 10Y-2Y spread has been negative for 12 months and the S&P 500 is 15% higher. Has the inversion “not worked”? a) Yes, it is a false indicator b) No – the inversion has a 6-18 month lag and first prices in cuts, which are positive for stocks. It only works out at the transition to phase 4, when earnings fall c) You should buy more

Correct: b) The cycle is a sequence, not a daily signal.

2. PMI New Orders - Inventories = -3.5. What does that mean? a) Companies are building inventories, production will rise b) Companies see fewer orders than the inventories they hold – they will cut production and orders to suppliers c) It does not matter

Correct: b) It is a classic signal of a slowing supply chain.

3. In which phase does the buy-the-dip strategy have the statistically worst result? a) In recovery b) In the transition from overheating to slowdown – the declines are the start of a longer downtrend in earnings c) It always works

Correct: b) Buying every dip in phase 3-4 is adding to a position that by definition is losing the support of earnings.

Frequently asked questions

Is the economic cycle in the US the same as in Poland? No. Poland is more dependent on the cycle in Germany, on EU funds and on the policy of the NBP (National Bank of Poland). The US sets the global cost of money, but the local cycle can differ by 2-3 quarters. You need two cockpits: a global one and a local one.

If the market leads the economy, why do I need macro data? To know whether the market is leading correctly. If the market is rising in anticipation of a recovery, but PMI keeps falling and claims are rising, the market may be correcting its expectations. Data verifies the market narrative.

How often should the cockpit be updated? Leading ones – weekly (spreads, the market). PMI, NFP, Claims – after publication. 20 minutes a week is enough to know which phase you are in. That is less than you lose on one trade without context.

---

This is not investment advice. The cycle is a risk filter, not fortune-telling the bottom. Measure the phase, adjust the position size.

---