Forensic Accounting Intelligence

Red Flags

Who Is The Next Enron?

Issue #3 · April 03, 2026 · Beneish M-Score • Altman Z-Score • Accruals Analysis • Governance Signals

⚠ Important Disclaimer: This newsletter applies academic forensic accounting models (Beneish M-Score, Altman Z-Score, Cash/Earnings Divergence Analysis, Governance Signals) to publicly available financial data. The presence of red flags is not evidence of fraud, wrongdoing, or impending financial distress. These are quantitative screens that identify statistical patterns warranting further due diligence. Many legitimate companies — particularly high-growth companies — exhibit elevated M-Scores or Z-Scores for entirely explainable, non-fraudulent reasons. This newsletter is for educational and research purposes only and does not constitute investment advice, a recommendation to sell or short any security, or an allegation of wrongdoing. Always consult a qualified financial advisor before making investment decisions. Past accuracy of these models does not guarantee future results.

Forensic Scoring Methodology — 100 Points Total ▼ click to expand

Four academic models combined into a composite fraud-risk score, using the same financial data already crawled by the persistent pipeline — no additional data collection. This panel appears only once per browser session.

Beneish M-Score (35 pts) — 1999 academic model detecting earnings manipulation via 8 financial ratios. The M-Score is a negative scale: values closer to zero are MORE suspicious. For example, −1.12 is closer to zero than −1.78, so −1.12 is the riskier score. Any score closer to zero than −1.78 classifies as a statistically likely manipulator. The model flagged Enron at this level in FY1997–1998, three years before the 2001 collapse.
M > −1.78 → 35 pts | M > −2.22 → 18 pts (grey zone)

Cash vs Earnings Divergence (25 pts) — Accruals ratio = (Net Income − Operating Cash Flow) / Total Assets. In every major accounting fraud, earnings were positive while cash flows lagged severely. "You can fake earnings, not cash." Multi-year pattern analysis (years with NI>0 but OCF<0) is the Luckin Coffee / Wirecard signature.
Accruals > 0.10 + NI>0/OCF<0 pattern → 25 pts

Altman Z-Score (20 pts) — 1968 bankruptcy prediction model. Five-factor formula incorporating working capital, retained earnings, EBIT, market cap vs debt, and revenue efficiency. Fraudulent accounting and financial collapse are often correlated.
Z < 1.81 → 20 pts (distress) | Z < 2.99 → 10 pts (grey zone)

Governance Signals (20 pts) — Short interest >20% of float (market-level fraud suspicion), minimal insider ownership, goodwill >40% of assets (acquisition risk / hidden write-offs), extreme beta.
Short >20% → up to 6 pts | Goodwill >40% → 5 pts | Low insider ownership → up to 4 pts

⚠ High-Growth False Positives: Fast-growing companies (high SGI, rising DSRI, negative OCF) will naturally score high on these models without any fraud. Look for the HIGH-GROWTH CAUTION notes above each stock's triggered flags. Universe: Russell 1000 (≥$1B market cap), screened weekly. Top 3 highest-scoring stocks featured. 6-month exclusion window prevents repeat coverage.

This Issue

MARA — MARA Holdings, Inc. (CRITICAL, 65/100)
WULF — TeraWulf Inc. (HIGH, 62/100)

#1 Highest Risk — Financial Services

MARA — MARA Holdings, Inc.

CRITICAL Red Flags Score: 65.0 / 100

$7.96

Market Cap: $3.0B

Capital Markets

About MARA Holdings, Inc.

MARA Holdings, Inc. operates as a digital asset technology company in the United States and Europe. It also provides technology solutions to optimize data center operations, such as liquid immersion cooling and firmware for bitcoin miners. The company was formerly known as Marathon Digital Holdings, Inc. and changed its name to MARA Holdings, Inc. in August 2024. MARA Holdings, Inc. was incorporated in 2010 and is based in Hallandale Beach, Florida.

📌 5-Year Weekly HLOC Chart: Weekly candlesticks showing High, Low, Open, Close. Green candles = weekly close above open; red = close below open. Use this to contextualise the long-term price trend over the period covered by the forensic analysis below.

📌 Revenue / Net Income / Operating Cash Flow: Quarterly where available, otherwise annual. Blue bars = Revenue. Green = positive Net Income or Cash Flow. Red = negative Net Income. Amber = negative Cash Flow. A persistent gap between net income and operating cash flow is the core accruals signal.

📌 Revenue vs Receivables: If accounts receivable grow faster than revenue, the company is booking sales before the cash arrives — the classic channel-stuffing or early-recognition pattern. Enron's AR grew 240% in the two years before collapse while revenue grew only 40%.

📌 Earnings vs Cash Flow: Shaded red columns indicate years where net income was positive but operating cash flow was negative — the Luckin Coffee and Wirecard signature. "You can fake earnings, not cash." A persistent gap (accruals) is the #1 quantitative fraud predictor.

📌 AR/Revenue Ratio (DSRI Trend): A rising ratio means receivables are accumulating faster than sales — a key Beneish Days Sales Receivable Index (DSRI) signal. Values consistently above 0.15–0.20 for most sectors warrant scrutiny. The Beneish model flags companies where DSRI exceeds 1.46× the prior year.

📌 Accruals Ratio (TATA) by Year: Total Accruals to Total Assets = (Net Income − Operating Cash Flow) / Total Assets. Red bars (>0.05) indicate earnings quality concern; bars above 0.10 are a strong fraud signal. Negative TATA (green) is healthy — cash flow exceeds reported earnings.

📌 Gross Margin Trend (GMI): Sustained margin compression creates pressure to manipulate reported earnings to meet analyst expectations. The Beneish Gross Margin Index (GMI) flags when prior-year margins were significantly better than the current year. Red bars = margin declined; green = improved.

📌 Altman Z-Score: Five-factor bankruptcy prediction model. Red zone (<1.81) has historically produced significant failure rates. Companies in financial distress have strong incentives to manipulate accounting — distress and fraud are correlated. Grey zone (1.81–2.99) warrants monitoring.

📌 Score Decomposition: The composite Red Flags Score (0–100) broken down by contributing model. Beneish M-Score contributes up to 35 pts, Cash Divergence 25 pts, Altman Z-Score 20 pts, Governance Signals 20 pts. Scores ≥65 = CRITICAL, ≥45 = HIGH, ≥25 = ELEVATED.

Forensic Analysis

FORENSIC SUMMARY — MARA Holdings, Inc. (MARA)

Composite Red Flags Score: 65.0/100 (CRITICAL)

⚠️ IMPORTANT CONTEXT: This company has growing revenues (+60%/yr). Several forensic ratios (Beneish M-Score, accruals, DSRI) are known to flag high-growth companies even when no manipulation is present. The signals below should be interpreted with this growth context in mind.

Beneish M-Score: -0.40 (BREACHES manipulation threshold (closer to zero than −1.78))

Altman Z-Score: 1.19 (DISTRESS ZONE)

Accruals Ratio: 0.000

Key findings:

• NOTE: Altman Z-Score was calibrated on US manufacturing firms and is less reliable for financial services companies. Interpret the Z-Score with caution for this stock.

• ⚠️ HIGH-GROWTH CAUTION: Revenue CAGR 60%/yr over the measured period. Many forensic signals (high SGI, rising DSRI, negative OCF) are EXPECTED in high-growth companies and do NOT necessarily indicate fraud. Verify that cash burn is funding genuine growth (R&D, sales expansion, customer acquisition) rather than masking accounting manipulation.

• Beneish M-Score -0.40 — breaches the −1.78 manipulation threshold. The M-Score is a negative scale where values closer to zero are MORE suspicious (e.g. −1.12 is closer to zero than −1.78, therefore riskier). Scores closer to zero than −1.78 classify as statistically likely earnings manipulators — the same model flagged Enron at this level in FY1997–1998, three years before the 2001 collapse. [35/35 pts]

• TATA 0.179 — accruals represent 17.9% of total assets (earnings quality concern)

• SGI 1.69 — revenue grew 69% year-on-year (rapid growth can mask channel stuffing)

Triggered Forensic Flags

❌NOTE: Altman Z-Score was calibrated on US manufacturing firms and is less reliable for financial services companies. Interpret the Z-Score with caution for this stock.

⚠️⚠️ HIGH-GROWTH CAUTION: Revenue CAGR 60%/yr over the measured period. Many forensic signals (high SGI, rising DSRI, negative OCF) are EXPECTED in high-growth companies and do NOT necessarily indicate fraud. Verify that cash burn is funding genuine growth (R&D, sales expansion, customer acquisition) rather than masking accounting manipulation.

❌Beneish M-Score -0.40 — breaches the −1.78 manipulation threshold. The M-Score is a negative scale where values closer to zero are MORE suspicious (e.g. −1.12 is closer to zero than −1.78, therefore riskier). Scores closer to zero than −1.78 classify as statistically likely earnings manipulators — the same model flagged Enron at this level in FY1997–1998, three years before the 2001 collapse. [35/35 pts]

❌TATA 0.179 — accruals represent 17.9% of total assets (earnings quality concern)

❌SGI 1.69 — revenue grew 69% year-on-year (rapid growth can mask channel stuffing)

❌Altman Z-Score 1.19 in DISTRESS ZONE (<1.81) — statistically high probability of financial distress. Companies scoring <1.81 have historically failed at significantly elevated rates.

❌Short interest 38.8% of float — elevated market scepticism (threshold: 20%)

❌Insider ownership 0.62% — below 1%

❌Beta 5.53 — high market volatility may reflect instability

Key Metrics

Beneish M-Score -0.40 ❌	Altman Z-Score 1.19 ❌	Accruals Ratio 0.000 ✅	Short Interest 38.8%
Beneish Score 35/35 pts	Altman Score 20/20 pts	Cash Div Score 0/25 pts	Gov Score 10/20 pts

Ticker $MARA is available to trade on eToro, where it may be available for Puts or a Short position.

#2 Highest Risk — Financial Services

WULF — TeraWulf Inc.

HIGH Red Flags Score: 62.4 / 100

$14.67

Market Cap: $6.2B

Capital Markets

About TeraWulf Inc.

TeraWulf Inc., together with its subsidiaries, owns, develops, operates digital infrastructure in the United States. It also develops and operates bitcoin mining facilities for bitcoin mining and high-performance computing workloads, leveraging clean, cost-effective, and reliable energy. The company was founded in 2021 and is headquartered in Easton, Maryland.

Forensic Analysis

FORENSIC SUMMARY — TeraWulf Inc. (WULF)

Composite Red Flags Score: 62.4/100 (HIGH)

⚠️ IMPORTANT CONTEXT: This company has growing revenues (+124%/yr). Several forensic ratios (Beneish M-Score, accruals, DSRI) are known to flag high-growth companies even when no manipulation is present. The signals below should be interpreted with this growth context in mind.

Beneish M-Score: 61.51 (BREACHES manipulation threshold (closer to zero than −1.78))

Altman Z-Score: 0.56 (DISTRESS ZONE)

Accruals Ratio: 0.000

Key findings:

• NOTE: Altman Z-Score was calibrated on US manufacturing firms and is less reliable for financial services companies. Interpret the Z-Score with caution for this stock.

• ⚠️ HIGH-GROWTH CAUTION: Revenue CAGR 124%/yr over the measured period. Many forensic signals (high SGI, rising DSRI, negative OCF) are EXPECTED in high-growth companies and do NOT necessarily indicate fraud. Verify that cash burn is funding genuine growth (R&D, sales expansion, customer acquisition) rather than masking accounting manipulation.

• ⚠️ RECEIVABLES CONTEXT: Rising AR/Revenue ratio (DSRI 2.12) is common in enterprise software and B2B companies shifting to larger contracts with longer payment terms. Compare AR days to peers — channel stuffing is only confirmed if AR days exceed industry norms by a substantial margin.

• ⚠️ PRE-PROFITABILITY CONTEXT: Company is currently loss-making. Pre-profitability companies often show negative OCF and high accruals simply due to investment phase economics. Altman Z-Score and accruals models are less reliable for companies not yet generating sustainable earnings. Focus on cash runway and revenue growth quality instead.

• Beneish M-Score 61.51 — breaches the −1.78 manipulation threshold. The M-Score is a negative scale where values closer to zero are MORE suspicious (e.g. −1.12 is closer to zero than −1.78, therefore riskier). Scores closer to zero than −1.78 classify as statistically likely earnings manipulators — the same model flagged Enron at this level in FY1997–1998, three years before the 2001 collapse. [35/35 pts]

Triggered Forensic Flags

❌NOTE: Altman Z-Score was calibrated on US manufacturing firms and is less reliable for financial services companies. Interpret the Z-Score with caution for this stock.

⚠️⚠️ HIGH-GROWTH CAUTION: Revenue CAGR 124%/yr over the measured period. Many forensic signals (high SGI, rising DSRI, negative OCF) are EXPECTED in high-growth companies and do NOT necessarily indicate fraud. Verify that cash burn is funding genuine growth (R&D, sales expansion, customer acquisition) rather than masking accounting manipulation.

⚠️⚠️ RECEIVABLES CONTEXT: Rising AR/Revenue ratio (DSRI 2.12) is common in enterprise software and B2B companies shifting to larger contracts with longer payment terms. Compare AR days to peers — channel stuffing is only confirmed if AR days exceed industry norms by a substantial margin.

⚠️⚠️ PRE-PROFITABILITY CONTEXT: Company is currently loss-making. Pre-profitability companies often show negative OCF and high accruals simply due to investment phase economics. Altman Z-Score and accruals models are less reliable for companies not yet generating sustainable earnings. Focus on cash runway and revenue growth quality instead.

❌Beneish M-Score 61.51 — breaches the −1.78 manipulation threshold. The M-Score is a negative scale where values closer to zero are MORE suspicious (e.g. −1.12 is closer to zero than −1.78, therefore riskier). Scores closer to zero than −1.78 classify as statistically likely earnings manipulators — the same model flagged Enron at this level in FY1997–1998, three years before the 2001 collapse. [35/35 pts]

❌DSRI 2.12 — accounts receivable growing 112% faster than revenue (revenue recognition risk)

❌Altman Z-Score 0.56 in DISTRESS ZONE (<1.81) — statistically high probability of financial distress. Companies scoring <1.81 have historically failed at significantly elevated rates.

❌Short interest 27.0% of float — elevated market scepticism (threshold: 20%)

❌Beta 4.30 — high market volatility may reflect instability

Key Metrics

Beneish M-Score 61.51 ❌	Altman Z-Score 0.56 ❌	Accruals Ratio 0.000 ✅	Short Interest 27.0%
Beneish Score 35/35 pts	Altman Score 20/20 pts	Cash Div Score 0/25 pts	Gov Score 7/20 pts

Ticker $WULF is available to trade on eToro, where it may be available for Puts or a Short position.

Red Flags by RoboMacro — Forensic Accounting Intelligence | Issue #3 | April 03, 2026

Models: Beneish (1999), Altman (1968). Data: Yahoo Finance, SEC EDGAR, Financial Modeling Prep.

This publication is for educational purposes only. Not investment advice. Presence of red flags does not constitute an allegation of fraud or wrongdoing. Do your own due diligence.

Cartographic Modernism