The Top-Down EIC Framework and Why SA 520 Secretly Runs on It

A company's fortunes are never entirely its own. A jewellery retailer's revenue trend is shaped by gold prices and rural income; a mining company's realisation is shaped by global commodity cycles; a real estate developer's collections are shaped by home-loan rates. Fundamental analysis has always recognised this by working top-down - economy first, industry second, company last because each layer sets the boundary conditions for the one below it.

What's less discussed is that this is exactly the discipline SA 520 (Analytical Procedures) demands from an auditor, whether at the risk-assessment stage or as a substantive procedure. SA 520 requires an auditor to form an independent expectation of a balance or ratio, evaluate the plausibility of relationships among data - financial and non-financial - and investigate any variance beyond an acceptable threshold. An expectation built only from last year's number plus a flat growth rate is weak. An expectation built top-down, layer by layer, is defensible - because at each step you're asking the standard's own question: is this relationship plausible, and why?

The Three Layers

Economy. This is the widest lens: GDP growth, interest rate trajectory, inflation, credit growth, currency movements, and the broader business cycle. It doesn't predict any one company's number, but it sets the ceiling and floor. A 6.5% GDP print with tightening credit growth tells you consumption-linked and capital-intensive sectors face different headwinds - before you've opened a single ledger.

Industry. The economy filters down unevenly. Within the same GDP print, cement may be capacity-constrained while IT services face a demand slowdown. This layer is where sector-specific drivers live: regulatory change, input-cost pass-through, competitive intensity, capacity utilisation, and how peers in the same audit portfolio are actually performing this quarter.

Company. Only at this final layer do you bring in the entity's own history, budget, segment mix, and related-party arrangements. This is also where the interesting audit evidence sits - because whatever the economy and industry don't explain, the company must.

Where the Two Frameworks Meet

Read top to bottom, that table is a residual: once the economy explains what it can and the industry explains what it can, whatever variance remains has to be explained by the entity itself - and that residual is precisely where SA 520 directs your investigation.

A Worked Shape

Take an auditee's revenue growing 18% year-on-year. Taken alone, that number is neither comforting nor alarming - it's just a number.

• Economy layer: nominal GDP growth for the year was ~11%, so 18% growth already sits above the macro base rate - worth a note, not yet a red flag.
• Industry layer: if the sector itself grew 20% on volume and price tailwinds, the entity's 18% is actually below-industry - the expectation shifts from "is this overstated?" to "why is this entity underperforming its sector?"
• Company layer: a new plant commissioned mid-year, or a large order booked in Q4, can fully explain the gap - and that explanation should be corroborated, not simply accepted.

Notice what happened: the same 18% produced three different questions depending on which layer you were standing in. That's the value of working top-down - it stops the auditor from anchoring on the entity's own trend line as the only benchmark, which is the single most common weakness in analytical procedures that don't hold up to review.

Why This Matters Beyond the Workpaper

An expectation with no macro or industry grounding invites the obvious challenge from a reviewer: why did you expect this number? "Because it's close to last year" is not a plausible relationship under SA 520 - it's an absence of one. Layering economy and industry data into the expectation-setting step converts analytical procedures from a mechanical variance check into genuine audit evidence, and it does so using data that, for most engagements, is a search away.

The same discipline also travels well beyond audit - it's how equity analysts frame a stock, how credit teams frame a borrower, and how any of us should frame a business we're trying to understand quickly. Start wide, narrow with purpose, and let whatever's left over be the thing you actually investigate.