PTHow long does a company's money sit idle?
Between buying, storing, selling and getting paid, a company's money spends weeks tied up in day-to-day operations. This is working capital. The thesis followed Brazilian companies listed on the stock exchange (B3) from 2015 to 2024 to see how they managed this money before and after the pandemic, and whether it affected profits. Explained in 12 screens.
The money cycle: the company buys, stores, sells and gets paid. The red cube is the money going around. Illustrative image.
The working capital clock counts days
The company pays its supplier, keeps the goods in inventory, sells on credit and only later collects from the customer. The days its money stays tied up along this path make up the cash conversion cycle.
Cycle = days of inventory + days to collect − days to pay the supplier. Move the periods and watch the cycle change.
Two strategies. According to the thesis, conservative management keeps more inventory and more sales on credit (more safety, lower expected return); aggressive management trims these days and stretches supplier terms (higher expected return, more risk). The averages are from the 92 companies in article 3 (Table 3.2), rounded.
One question, three articles, two periods
How did Brazilian companies listed on B3 manage their working capital before and after the pandemic? The thesis answers in three steps.
After 2020, researchers paid more attention to working capital
The author combined the two largest databases of scientific articles in the field (Scopus and Web of Science) and kept 168 open-access articles on working capital, published from 2015 to 2024.
Publications per year jumped in 2020, the year the pandemic began, and did not return to their previous level.
Values from Figure 1.1 of the thesis; 2024 counts only up to October. The totals before and after 2020 are our count. Search funnel (Table 1.1): 316 records → 184 after the field, type and language filters → 168 without the 15 duplicates across the two databases.
Many wrote once; hardly anyone wrote several times
This is what Lotka's Law predicts: a few researchers publish a lot and many publish little. Something similar happens with journals (Bradford's Law): a small core concentrates a large share of the articles.
After 2020, money stayed longer in inventory
Comparing the averages of the 92 companies, the cash cycle went from 94 to 103 days. Inventory drove it: 25.7% more days sitting idle.
Companies also took longer to pay suppliers and collected from customers slightly faster. For the thesis, this was a more flexible (conservative) management, with more slack to get through the uncertainty. Tap a pair of blocks.
Averages from Table 3.2 (extreme values not treated). Values vary widely across companies: the standard deviation of the cycle exceeds 200 days. In article 2, with 107 companies, the average cycle went from 91 to 98 days, and the median from 33 to 35 days (Table 2.3).
Strong cash flow, large size and growing sales: a shorter cycle
Article 2 measured what goes along with a shorter or longer cycle. In both periods, companies that generate more cash from operations, that are larger and whose sales grow faster each quarter had shorter cycles.
The thesis measured working capital in two ways: in days (the cycle) and in money (the working capital requirement: how much of total assets is tied up in inventory and receivables, minus what suppliers finance).
See the full-model coefficients as a table
Tables 2.4 and 2.5 (fixed effects). Operating cash flow and return on assets move closely together, so the thesis also estimated models with one at a time. Profit and cycle: the negative effect only appears in the model without cash flow. The cycle model explains 19% (before) and 25% (after) of each company's variation over time; the capital requirement model, only 2% to 3%.
The pandemic changed the strength, not the direction
The factors that shortened the cycle before 2020 kept shortening it afterwards. What changed was the size of the effect, and debt, which only started to matter after the pandemic.
Choose a factor and see how many fewer days of cycle go with it in each period, starting from the average cycle.
Our calculation. Starts from the average cycle of the 107 companies (91 and 98 days, Table 2.3) and applies the full-model coefficients (Table 2.4), holding everything else constant. Only statistically significant effects are included. Shows association, not cause.
Shorter cycle, slightly higher profit, before and after the pandemic
Among the 92 companies, those that kept their money idle for fewer days had a higher return on assets (ROA). This holds in both periods; after 2020, the relationship got stronger.
For the thesis, this argues for aggressive working capital management. But the effect per day is small, as the author himself points out. Shorten the cycle and see.
Our calculation. Starts from the average ROA (4% before, 5% after; Table 3.2) and applies the cycle coefficient (−0.0000193** per day before; −0.0000291*** after; Tables 3.3 and 3.4). ROA is a fraction in the thesis: 0.0000193 per day = 0.00193 percentage point. Shows association, not cause.
After the pandemic, the time given to customers to pay weighed three times more
The thesis also tested each piece of the cycle. Before 2020, 30 more days to collect from customers came with an ROA 0.38 point lower; afterwards, 1.17 points lower.
Inventory days weighed little (and, after 2020, only with weak significance, at 10%). And the time taken to pay suppliers? It appears on the next screen.
Our calculation from the coefficients in Tables 3.3 and 3.4 (random effects), converted into percentage points of ROA. Each bar comes from a separate model. Stars: significance at 10% (*), 5% (**) and 1% (***).
Not everything followed the aggressive-management recipe
Place your bet and flip the card.
Payment terms are day-to-day decisions, and they add up
Financial managers
Tracking inventory, collection and payment periods one by one helps explain profit: a shorter cycle came with a higher ROA in both periods.
In times of crisis
The relationships did not change direction with the pandemic; they changed intensity. After 2020, the time given to customers weighed more on profit.
Indebted companies
After the pandemic, more indebted companies had shorter cycles, which the thesis reads as greater care for liquidity amid high interest rates.
Researchers
Brazil barely appears in world research on the topic and had no international partnerships in the period; the thesis points to the scarcity of studies on Brazilian companies.
What this research does not say
- Few companies, concentrated in some sectors: 107 and 92 listed companies, almost half of them in manufacturing. · according to the author
- Chosen time windows and variables. Other periods, sectors and small and medium-sized companies may show something else. · according to the author
- Only open-access articles, from two databases, in the survey of article 1. · according to the author
- Small effect per day ("reduced practical significance", in the author's words) and association, not proof of cause. · our note
Credits
IORIO, Fabio Roberto. Gestão do capital de giro de empresas brasileiras listadas na B3: uma análise dos contextos pré e pós pandemia. Tese (Doutorado em Administração de Empresas) — Universidade Presbiteriana Mackenzie, São Paulo, 2025. Orientador: Prof. Dr. Wilson Toshiro Nakamura.
Research supported by the Mackenzie Presbyterian Institute (full tuition and fee waiver). Data from Economática, Scopus and Web of Science. This version simplifies the work for the general public; the formulas and full tables are in the original text (in Portuguese).