WhitmanTrading

SG&A: Read It as a Share of Sales

SG&A - selling, general and administrative expenses - is the cost of running a company that is not attributable to producing what it sold. Most of the total is fixed in the short term, which is why a fall in revenue produces a larger proportional fall in profit.

How it works

A labelled statement diagram showing revenue less cost of goods sold giving gross profit, then less selling general and administrative expenses giving operating income. The headline reads: The cost of selling and running the company, not of making it.
The cost of selling and running the company, not of making it. Illustrative figures - not a real company.

SG&A sits between gross profit and operating income. The costs of making the product are already gone; this line is everything else needed to run the business and sell what it makes.

A labelled statement diagram splitting the line into selling, general and administrative components. The headline reads: Three different kinds of cost share one line.
Three different kinds of cost share one line. Illustrative figures - not a real company.

Three distinct categories share one label. Selling covers sales staff, commissions and marketing. General and administrative covers head office, finance, legal and the rest of the machinery. Only the first has any natural relationship to volume.

A labelled statement diagram comparing cost of goods sold with overheads. The headline reads: Cost of goods scales with volume; this line does not.
Cost of goods scales with volume; this line does not. Illustrative figures - not a real company.

The contrast with cost of goods sold is the point. Sell half as much and the cost of goods roughly halves; the head office does not.

Why it drives operating leverage

A labelled statement diagram comparing the fixed and variable portions of the overhead line. The headline reads: Most of it does not fall when revenue falls.
Most of it does not fall when revenue falls. Illustrative figures - not a real company.

Most of the line is fixed in the short term. Leases, salaries and systems continue whatever happens to sales, and cutting them takes quarters rather than weeks.

A labelled statement diagram showing revenue down ten per cent with cost of goods falling proportionally and overheads unchanged, giving a much larger fall in operating income. The headline reads: Which is why profit moves further than sales do.
Which is why profit moves further than sales do. Illustrative figures - not a real company.

That fixity is operating leverage. In the illustration a 10% fall in revenue produces a 23% fall in operating income, because 900 of overheads did not move while everything above it did.

The same mechanism works upward. A 10% rise in revenue produces a disproportionately larger rise in profit, which is why operationally geared businesses are volatile in both directions and why their earnings are harder to forecast than their sales.

A labelled statement diagram comparing selling costs tied to volume with administrative costs tied to nothing. The headline reads: Selling cost should rise with sales; admin should not.
Selling cost should rise with sales; admin should not. Illustrative figures - not a real company.

The two halves should behave differently. Selling costs rising with revenue is normal; administrative costs rising with revenue means the business is not gaining any efficiency from being larger.

In practice: read the ratio

A labelled statement diagram showing the overhead ratio across three years at twenty-two, twenty-three and twenty-six per cent. The headline reads: Read it as a share of revenue, never as a number.
Read it as a share of revenue, never as a number. Illustrative figures - not a real company.

The absolute figure is meaningless without the revenue beside it. Overheads as a percentage of sales, tracked across years, is the version that carries information.

A labelled statement diagram comparing last year's overheads with a higher figure this year against unchanged revenue. The headline reads: A rising share on flat sales is the warning.
A rising share on flat sales is the warning. Illustrative figures - not a real company.

A rising ratio on flat revenue is the signal to investigate. Costs growing without sales growing is either an investment that has not paid off yet or a business losing control of itself, and the accounts alone will not say which.

A labelled statement diagram splitting the overhead total into cash overheads and share-based pay. The headline reads: Share-based pay sits inside it and uses no cash.
Share-based pay sits inside it and uses no cash. Illustrative figures - not a real company.

Share-based compensation is inside this line and uses no cash. It is a real cost to shareholders through dilution, and it is the reason a company’s overheads can exceed the cash it actually spent on them.

A labelled statement diagram showing overheads before and after cuts with the saving as the result. The headline reads: It is the first line management cuts, and the easiest to cut badly.
It is the first line management cuts, and the easiest to cut badly. Illustrative figures - not a real company.

It is the first line cut under pressure. Marketing and headcount are reducible quickly, which is exactly why cuts here can raise this year’s profit while removing next year’s growth.

A labelled statement diagram comparing a company reporting one combined line with one reporting three separate lines. The headline reads: And some companies split it out while others do not.
And some companies split it out while others do not. Illustrative figures - not a real company.

Presentation varies between companies. Some report selling, general and administrative separately; some combine everything; some move research spending in or out. Comparing two companies means checking what each one put in the line.

A labelled statement diagram showing the best, median and worst overhead ratio across five years. The headline reads: Track the ratio for five years, not the figure for one.
Track the ratio for five years, not the figure for one. Illustrative figures - not a real company.

Five years of the ratio is the useful view. One year has no trend in it, and the trend is what distinguishes a business getting more efficient from one that had a good quarter.

One split inside the line is worth reconstructing even though companies rarely publish it: what is growing the business against what is running it. Sales and marketing spending buys future revenue; finance, legal and head office keep the lights on. A company where the first is growing and the second is flat is scaling; one where the second is growing is adding weight.

The segment note and the cash flow statement usually contain enough to estimate the split. Headcount disclosures, marketing spend where it is given, and share-based pay by function all help. Even a rough split changes the reading, because a rising overhead ratio driven entirely by sales investment is a different company from one driven by administration.

And the ratio has a natural floor that varies by business model. A software company will never get its overheads down to a retailer’s percentage, because the cost of acquiring a customer is most of what it spends. Compare a company with its own history first and its industry second — those two comparisons answer different questions, and the cross-industry one answers almost nothing.

What SG&A is not

It is not the cost of the product. That sits above, in cost of goods sold.

It is not all fixed. Selling costs move with volume; the rest largely does not.

It is not comparable across companies without checking. Definitions vary.

And it is not all cash. Share-based pay is inside it.

When it fails as a signal

A rising ratio can be a deliberate investment. A company building a sales force ahead of growth reports exactly the same pattern as one losing cost control, and only the following two years distinguish them.

A second failure is a falling ratio achieved by cutting. Overheads down and revenue flat looks like efficiency for a year and frequently shows up as lost growth afterwards.

A third is comparing across industries. A software company’s overheads are a much larger share of revenue than a retailer’s, and neither figure is wrong.

A fourth is ignoring what moved between lines. Reclassifying a cost from cost of goods to overheads changes gross margin and this ratio simultaneously, with no operational change at all.

And a fifth is reading it without revenue. The number alone says nothing.

The original data

Of the 31,760 trading and investing videos in this site’s corpus, 0 have “income statement” in the title and 3 have “financial statements” at a median of 1,065,893 views. “Cash flow” returns 17 at a median of 67,134 across 14 channels, and “valuation” returns 9 at a median of 17,868. The counts are in research/corpus-coverage.json, produced by site/measure_corpus.py.

The figures in the diagrams on this page are illustrative and chosen to make the leverage visible. The 10% revenue fall producing a 23% operating income fall is the arithmetic of 900 in fixed overheads against 1,600 of gross profit, and it is the single most useful thing this line explains.

The habit worth adopting is one division, five times. Overheads over revenue for each of the last five years, written in a row. A ratio drifting upward is the earliest warning most sets of accounts give you, and it appears well before it reaches the profit figure anybody quotes.

Operating expenses is the wider category this belongs to. Operating income is what remains after it. And income statement is the statement it sits inside.

What I actually do

The ratio is the only version of this line I look at. An absolute figure tells me nothing without knowing the size of the business, and the trend in the ratio over five years has told me more about how a company is being run than almost any other single number.

— Michael Whitman

This page is educational, not financial advice. Test every idea on your own charts before risking money.