Stock Showdown: AGC vs. Nippon Electric Glass vs. Hoya
TSE: 5201 vs. 5214 vs. 7741 | Japanese Glass Substrate Triad | March 2026
Three Japanese companies. All make glass. All serve the display and semiconductor industries. All trade on the Tokyo Stock Exchange. That is where the similarities end.
AGC is a diversified materials conglomerate that happens to make display glass. NEG is the purest play on the display glass oligopoly. Hoya is a precision optics juggernaut that dominates EUV mask blanks and HDD glass substrates. Buying "Japanese glass substrate companies" without understanding these differences is like buying "American tech" and not caring whether you get Apple or Dell.
This report compares the three across every dimension that matters, picks a winner, and tells you what to do with your money.
1. At-a-Glance Snapshot
| Metric | AGC (5201.T) | NEG (5214.T) | Hoya (7741.T) |
|---|---|---|---|
| Company Name | AGC Inc. | Nippon Electric Glass Co., Ltd. | HOYA Corporation |
| Sector | Materials / Glass & Chemicals | Materials / Specialty Glass | Technology / Precision Optics |
| Market Cap | ~1.32T yen (~$8.8B) | ~460B yen (~$3.1B) | ~9.3T yen (~$62B) |
| Enterprise Value | ~1.5T yen | ~448B yen (net cash) | ~8.8T yen |
| Price (Mar 17, 2026) | ~4,500 yen | ~6,250 yen | ~23,700 yen |
| 52-Week Range | 3,870 -- 6,959 | 2,999 -- 7,185 | 14,345 -- 29,590 |
| Revenue (LTM) | 2,059B yen (FY2025) | 311B yen (FY2025) | 866B yen (FY24, Mar 2025) |
| Operating Margin | 6.2% | 11.0% | 29.5% |
| Net Margin | 3.4% | 9.5% | 23.3% |
| ROE | ~5% | ~10% | ~21% |
| P/E (TTM) | ~21x | ~16x | ~32x |
| P/B | 0.91x | 0.94x | ~8.9x |
| Dividend Yield | ~4.3% | ~2.6% | ~0.9% |
| Employees | ~52,900 | ~5,500 | ~37,000+ |
(Source: Yahoo Finance, StockAnalysis, MarketScreener, Investing.com, company IR materials. Prices and multiples reflect latest available data as of mid-March 2026.)
The numbers tell the story immediately. Hoya generates 5x the operating margin of AGC on half the revenue. NEG is the smallest but has the purest exposure to display glass. AGC is the cheapest on P/B but generates the lowest returns on equity. You get what you pay for in this market, and the valuation spread reflects genuine business quality differences.
2. Business Model Comparison
AGC is a sprawling materials conglomerate. It makes architectural glass for buildings (~25% of revenue), automotive glass (~20%), display glass substrates (~10%), electronic materials including EUV mask blanks (~8%), chlor-alkali and fluorochemicals (~27%), and biopharmaceutical CDMO services (~7%). The company operates ~200 subsidiaries across 30+ countries. Think of AGC as a department store of materials: broad selection, something for everyone, but no single counter dominates.
NEG is the display glass specialist. About 56% of revenue comes from electronics and information technology (primarily display glass substrates for LCD/OLED panels), and 44% from functional materials (glass fiber for automotive and PCBs, plus specialty glass). NEG is the price competitor in the display glass oligopoly -- typically 5-10% cheaper than Corning, with plants strategically located near Asian panel makers. The company also has an emerging semiconductor glass substrate business (GC Core) that is pre-revenue but potentially transformative.
Hoya is a precision optics and glass company with two engines. The Life Care segment (~64% of revenue) makes eyeglass lenses (#2 globally), contact lenses, endoscopes (#3 globally), and intraocular lenses (#3 globally). The Information Technology segment (~36% of revenue) makes EUV mask blanks (dominant 75%+ share), HDD glass substrates (~100% consumer, ~40% nearline), and FPD photomasks. The IT segment generated 54.1% operating margins in FY2024 -- extraordinary for any business, let alone a manufacturing one.
| Dimension | AGC | NEG | Hoya |
|---|---|---|---|
| Revenue Model | Product sales across 8+ segments. Mix of commodity (glass, chemicals) and specialty (mask blanks, CDMO). ~1/3 recurring. | Product sales in 2 segments. Predominantly transactional, but long furnace campaigns (5-8 years) create de facto supply lock-in. | Mixed: consumable-like recurring (mask blanks, eyeglass lenses) + cyclical (HDD substrates, FPD masks). High-margin IT is the earnings engine. |
| Geographic Mix | Japan/Asia 64%, Europe 28%, Americas 8% | International 87%, Japan 13% | Life Care: 80%+ Europe/Americas/Japan. IT: 80%+ Asia (Taiwan, Korea, China, SE Asia). |
| Customer Concentration | Low (no single customer >5% of consolidated) | Moderate-High (top 5-6 panel makers = majority of display glass revenue) | Moderate in IT (TSMC, Samsung, Intel, Seagate); dispersed in Life Care (64,000+ optician accounts). |
| Competitive Moat | Narrow overall. Strong in niches (ETFE #1, display substrates #2, mask blanks). Weak in commodity glass. | Moderate. Oligopoly protection (3-player market, no new entrant in 20+ years), but #3 position and limited pricing power. | Wide. Dominant in EUV blanks (75%+), near-monopoly in consumer HDD glass (100%), 80+ years of optical IP (50,000+ glass compositions). |
| Capital Intensity | Very High. Float glass lines, chemical plants, CDMO facilities. Capex ~180B yen/year. | High. Continuous glass furnaces run 24/7 with 5-8 year campaigns. Capex ~30B yen/year. | Moderate. Mask blank production requires extreme precision but not massive physical scale. Capex ~50B yen/year. |
| Secular Tailwinds | Green buildings, EV glass content, EUV adoption, green hydrogen, biopharma outsourcing | Larger TV panels, automotive displays, EV lightweighting (glass fiber), 5G PCBs, semiconductor glass wafers | EUV layer-count growth, High-NA EUV, HAMR HDD transition, global myopia epidemic, aging populations |
Business Model Verdict
Hoya has the best business model by a wide margin. It operates at structural chokepoints where customers have no alternative, generates 30%+ operating margins, and returns 100% of free cash flow to shareholders. The consumable nature of mask blanks (they wear out) creates recurring demand that is far more predictable than AGC's commodity glass cycles or NEG's furnace utilization swings.
AGC has the worst business model for shareholders -- not because it is a bad company, but because the conglomerate structure dilutes the high-quality segments (electronic materials, fluorochemicals) with low-margin commodity businesses (architectural glass, basic chemicals) that drag returns on capital below the cost of capital.
NEG sits in between: a focused oligopolist with structural entry barriers, but one that is #3 in its core market with limited ability to gain share from Corning. The business model is sound but inherently cyclical -- when display demand drops, the high fixed-cost base turns profits into losses (as it did in FY2023).
3. Financial Health -- Side by Side
Income Statement
| Metric | AGC | NEG | Hoya |
|---|---|---|---|
| Revenue (LTM) | 2,059B yen (FY2025) | 311B yen (FY2025) | 866B yen (FY24, Mar 2025) |
| Revenue Growth (YoY) | -0.4% | +4.1% | +14% |
| Revenue Growth (3Y CAGR) | ~0.4% (essentially flat) | -1.4% (distorted by downcycle) | ~9.4% |
| FY2026E Revenue (mgmt guidance) | 2,200B yen (+6.9%) | 320B yen (+2.8%) | ~920-950B yen (+6-10%, my estimate) |
| Gross Margin | Not separately disclosed; OP margin is the metric | 25.7% (FY2025) | ~56% (estimated) |
| Operating Margin | 6.2% | 11.0% | 29.5% |
| Net Margin | 3.4% | 9.5% | 23.3% |
| Operating Income | 127B yen | 34.1B yen | 255.8B yen |
| FY2026E Operating Income (mgmt guidance) | 150B yen (+18%) | 33.0B yen (-3%) | ~270-285B yen (+6-11%, my estimate) |
| EPS (LTM) | 326 yen | 394 yen | ~584 yen (FY24) |
| EPS Growth Trajectory | Recovering from FY2024 net loss (-94B yen from impairments) | Massive recovery: from -294 yen (FY2023) to +394 yen (FY2025) | Consistent double-digit growth, augmented by buybacks |
AGC's income statement tells a story of stagnation. Revenue has been flat for three years, operating margins have been stuck at 6%, and the FY2024 net loss (from impairment charges, not operating deterioration) spooked the market. The FY2026 guidance of 150B yen operating income (+18%) would be the first real improvement in three years -- but management has to deliver it.
NEG's income statement tells a story of violent cyclicality. The company went from a 10.4B yen operating loss in FY2023 to 34.1B yen operating profit in FY2025. That is a 45B yen swing driven almost entirely by furnace utilization rates. When demand comes back, NEG prints money. When demand falls, it hemorrhages. The FY2026 guidance of 33B yen operating profit (slight decline) is conservative -- or management sees margin pressure ahead.
Hoya's income statement tells a story of consistent quality. Revenue grew 14% in FY2024, operating margins held near 30%, and the IT segment's 54.1% margin is world-class. The Life Care segment's margin declined 3 percentage points (PENTAX Medical drag), but that is the one area where execution has been weak.
Cash Flow & Balance Sheet
| Metric | AGC | NEG | Hoya |
|---|---|---|---|
| Operating Cash Flow (LTM) | 274B yen | ~55-60B yen (est.) | ~210-220B yen (est.) |
| Capex | ~178B yen | ~30B yen | ~50B yen |
| Free Cash Flow | 96B yen | ~25-30B yen (est.) | ~160-170B yen (est.) |
| FCF Margin | 4.7% | ~8-10% (est.) | ~19-20% |
| FCF Yield | ~7.3% | ~5-6% | ~1.8% |
| Net Debt / (Cash) | ~180B yen bonds outstanding; equity ratio 50.3% | Net cash (~23B yen) | Net cash (fortress) |
| Equity Ratio | 50.3% | 70.2% | Very high (debt/equity 0.04x) |
| Credit Rating | R&I AA / S&P A- / Moody's A2 | N/A (not rated, but fortress balance sheet) | N/A (no need for debt) |
| Dividend per Share | 210 yen (held since FY2021) | 160 yen (DOE 3% target) | ~160 yen (~0.9% yield) |
| Buyback Activity | Modest, keeps share count flat | Aggressive: 100B yen program (Nov 2023 -- Dec 2028), ~21% of market cap | Aggressive and consistent: 547B yen in 6 years; returns 100% of FCF |
| Total Shareholder Yield | ~4.5% (dividend + modest buyback) | ~11.8% (2.6% dividend + 9.2% buyback) | ~5-6% (dividend + buyback, but lower yield due to high valuation) |
The balance sheet comparison is straightforward: all three companies are in excellent financial health. None has any solvency risk. But the quality of cash flow generation is radically different.
AGC generates the most absolute free cash flow (96B yen) but on a massive revenue and asset base, yielding mediocre FCF margins of 4.7%. The 180B yen in outstanding bonds is manageable given investment-grade ratings, and the 210 yen dividend is well-covered.
NEG has the cleanest balance sheet -- net cash, 70% equity ratio, no debt concerns whatsoever. The 100B yen buyback program (representing over 20% of the current market cap, running through 2028) is genuinely aggressive and shareholder-friendly. But FCF will compress sharply in FY2026 due to elevated capex (~45B yen) for semiconductor glass and all-electric furnace investments.
Hoya generates by far the highest-quality free cash flow -- 160-170B yen at 19-20% FCF margins on a net cash balance sheet. The company has committed to returning 100% of FCF to shareholders through dividends and buybacks, and it has executed on that commitment year after year. The 547B yen in buybacks over six years speaks for itself.
4. Growth Comparison
| Growth Metric | AGC | NEG | Hoya |
|---|---|---|---|
| Revenue CAGR (3Y historical) | ~0.4% | -1.4% | ~9.4% |
| Revenue Growth (NTM, mgmt guidance) | +6.9% | +2.8% | +6-10% (my estimate) |
| Revenue CAGR (consensus, 2-year forward) | ~4-5% (consensus) | ~3-4% (consensus) | ~7-9% (consensus) |
| EPS Growth (NTM) | +63% (recovery from impairment year) | -22% (one-time FY2025 gains not repeating) | +12-15% (organic + buybacks) |
| Sustainable EPS Growth Rate (my estimate) | 5-8% | 3-6% | 12-15% |
| Fastest-Growing Segment | Electronic materials (~8-12% CAGR) | GC Core semiconductor glass (pre-revenue) | EUV mask blanks (~15% CAGR), HAMR glass substrates (~12% CAGR) |
| Biggest Growth Bet | AGC Biologics CDMO (payoff 2027-2028) | GC Core glass-ceramic substrates for chip packaging (mass production target 2028) | High-NA EUV blanks + HAMR nearline HDD transition |
The growth comparison is not even close. Hoya is growing revenue at nearly 10% a year organically, and buybacks add another 3-5% to EPS growth. AGC has been flat for three years and is hoping its FY2026 guidance marks an inflection. NEG is growing low-single-digits off a cyclical recovery and has no visible path to mid-single-digit growth without the GC Core semiconductor glass story materializing.
The quality of growth also matters. Hoya's growth is driven by structural tailwinds -- more EUV layers per chip (not cyclical, just physics), HAMR adoption requiring glass substrates (not optional, aluminum cannot handle the heat), and a global myopia epidemic that is not going away. These are as close to "guaranteed" growth drivers as exist in the industrial world.
AGC's growth depends on executing a portfolio transformation (shifting revenue mix toward electronic materials, fluorochemicals, and CDMO) that has been talked about for years but barely moved the needle on consolidated margins. The company's highest-growth segments represent only ~27% of revenue, so even if they grow 10%, the consolidated effect is muted.
NEG's growth story is the most binary. If the display glass cycle holds and GC Core semiconductor substrates reach mass production by 2028-2029, NEG could surprise. If the display cycle rolls over and GC Core stays niche, NEG reverts to low-growth mode. There is real optionality here, but it is unproven.
5. Valuation Comparison
Absolute Multiples
| Metric | AGC | NEG | Hoya |
|---|---|---|---|
| P/E (TTM) | ~21x | ~16x | ~32x |
| P/E (Forward, NTM) | ~12-14x (based on consensus EPS ~397 yen) | ~19-20x (based on FY2026E EPS ~306 yen) | ~28x (forward) |
| EV/EBITDA | ~4.2x | ~8.0x | ~19.5x |
| EV/Revenue | ~0.73x | ~1.44x | ~10.2x |
| P/B | 0.91x | 0.94x | ~8.9x |
| P/FCF | ~13.7x | ~16-20x | ~55x |
| FCF Yield | ~7.3% | ~5-6% | ~1.8% |
| Dividend Yield | ~4.3% | ~2.6% | ~0.9% |
(Sources: Yahoo Finance, StockAnalysis, Investing.com, Morningstar, company IR. Multiples reflect latest available data as of mid-March 2026. Forward estimates are analyst consensus where available, management guidance where noted, or my estimates where labeled.)
Relative to Own History (5-Year Range)
| Metric | AGC | NEG | Hoya |
|---|---|---|---|
| P/E range | ~10-25x | ~7-22x (when profitable) | ~25-45x |
| Current vs. range | Mid-range | Mid-range | Mid-to-upper range |
| P/B range | 0.5-1.1x | 0.41-0.94x | ~5-10x |
| Current vs. range | Mid-range (0.91x) | Upper end (0.94x) | Mid-range (~8.9x) |
| EV/EBITDA range | 3-8x | 4-12x | ~15-30x |
| Current vs. range | Low end (4.2x) | Mid-range (8.0x) | Mid-range (19.5x) |
AGC is cheap on absolute metrics (sub-1x P/B, 4.2x EV/EBITDA, 7%+ FCF yield) and mid-range relative to its own history. The discount reflects persistently low ROE (~5%) -- the market is correctly pricing in mediocre returns on capital.
NEG is trading at the highest P/B in 5 years (0.94x), which sounds alarming until you realize that is because (a) the stock doubled off its lows and (b) the company has been shrinking book value per share through buybacks. On forward P/E (19-20x), it looks reasonable given the mid-cycle earnings base.
Hoya is expensive on every absolute metric. At 32x trailing earnings, 19.5x EV/EBITDA, and 8.9x P/B, you are paying a premium that assumes sustained excellence. But Hoya has earned that premium consistently -- the stock has traded at 30-40x earnings for years because the business consistently delivers 20%+ ROE and 30%+ operating margins. The current multiple is in the middle of the historical range, not at extremes.
Growth-Adjusted Valuation
| Metric | AGC | NEG | Hoya |
|---|---|---|---|
| PEG Ratio (my estimate) | ~2.5-3x (12x forward P/E / 5% sustainable growth) | ~3-4x (19x forward P/E / 5% sustainable growth) | ~2.0-2.3x (28x forward P/E / 13% sustainable EPS growth) |
| EV/EBITDA / Growth | ~0.8x (4.2x / 5% growth) | ~1.6x (8x / 5% growth) | ~1.5x (19.5x / 13% growth) |
This is the most revealing comparison. On a growth-adjusted basis, AGC is the cheapest -- but that is because the market does not believe the growth will materialize. NEG is the most expensive on a growth-adjusted basis, reflecting the cyclical premium embedded in mid-cycle earnings. Hoya, despite its eye-watering absolute multiples, is actually reasonably valued on PEG because the growth rate is genuinely higher and more durable.
Valuation Verdict
If you are a value investor who buys low P/B and high dividend yields, AGC is your stock. If you want the purest cyclical recovery play with aggressive buybacks, NEG is your stock. If you want to own the highest-quality business and are willing to pay up for it, Hoya is your stock.
The market is broadly efficient here. The valuation spread reflects real differences in business quality, growth rates, and returns on capital. There is no obviously mispriced stock among the three.
6. Quality & Capital Allocation
| Dimension | AGC | NEG | Hoya |
|---|---|---|---|
| ROE (TTM) | ~5% | ~10% | ~21% |
| ROIC (TTM) | ~5-6% | ~5.1% | ~16-31% (range depends on definition) |
| ROIC vs. WACC | Roughly breakeven (ROIC ~5-6% vs. WACC ~6-7%). Not clearly creating value. | Roughly breakeven (ROIC ~5% vs. WACC ~6-7%). Barely covers cost of capital. | Massive spread (ROIC ~16-31% vs. WACC ~9-10%). Creating enormous economic value. |
| Moat Rating | Narrow | Moderate | Wide |
| Morningstar Moat | Not rated | Not rated | Wide Moat |
| Capital Allocation Grade | C+ (reliable dividend, mediocre ROE, passive buybacks) | B (aggressive buybacks, fortress balance sheet, but mediocre ROIC) | A- (100% FCF returned, disciplined bolt-on M&A, consistent execution; dinged for low insider ownership) |
| Buyback Effectiveness | Keeps share count flat. Not value-accretive -- buying at ~0.9x book is fine, but the company is not doing it aggressively enough. | Very effective. Buying back 5%+ of shares annually at or below book value. The 100B yen program is 21% of market cap. | Very effective. 547B yen in 6 years. Buying at high multiples but the ROE justifies it -- every yen returned earns 20%+ for shareholders. |
| Dividend Policy | 210 yen/share held since FY2021. Payout ratio ~64%. Well-covered by FCF. | 160 yen/share, DOE 3% target. Growing dividends. | ~160 yen/share. Low yield (~0.9%) but growing. Buybacks are the primary return mechanism. |
| Management Track Record | Competent, honest, conservative. Capital allocation is too passive. The "AGC plus" transformation has been slow. | Genuinely strong. The 100B yen buyback ahead of schedule, DOE 3% dividend, non-core asset divestitures (49B yen in FY2024). Run by career lifers who know the business. | World-class. The dual-engine model works, portfolio pruning is disciplined (PENTAX restructuring), and capital returns are relentless. |
The quality gap between Hoya and the other two is the single most important fact in this comparison. Hoya generates 4x the ROE of AGC and NEG. It earns well above its cost of capital, creating genuine economic value for shareholders every year. AGC and NEG are roughly breakeven on value creation -- they earn approximately their cost of capital, which means shareholders get a market-rate return but nothing more.
This is why Hoya trades at 9x book and the others trade at or below book. It is not the market being irrational -- it is the market correctly pricing the difference between a 20% ROE franchise and a 5% ROE conglomerate.
7. Risk Comparison
| Risk Factor | AGC | NEG | Hoya |
|---|---|---|---|
| Cyclical Risk | Medium. Diversified across cycles but exposure to construction, auto, and display. | High. Display glass demand is the dominant earnings driver, and it is inherently cyclical. Operating leverage cuts both ways -- violently. | Medium-Low. IT segment is cyclical (semiconductors), but Life Care provides countercyclical stability. Beta of 0.59 reflects this. |
| Competitive Risk | Medium. #2 in display glass (behind Corning's superior fusion-draw process), distant #2 in EUV blanks (behind Hoya). Neither position is dominant. | Medium. #3 in display glass with no realistic path to gaining share. Corning has the technology edge; NEG competes on cost. | Low. No competitor can match EUV blank quality. AGC is the only credible #2 but has not achieved parity on defect levels. Applied Materials is years away. |
| China Risk | Medium. Operations and revenue exposure in China. Regulatory uncertainty. | Medium-High. Chinese domestic glass substrate development (CNBM has produced Gen 8.6 OLED substrates) is a 5-10 year threat. NEG's largest growth market is at risk. | Medium. FPD operations in Chongqing; some IT revenue from China. Singapore as primary mask blank hub provides insulation. |
| FX Risk (Yen) | Medium. 72% revenue from outside Japan. Yen appreciation compresses margins. | High. 87% revenue from outside Japan. A move from 150/USD to 120/USD would meaningfully crush export margins. | Medium. IT segment (~80% Asia) has significant FX exposure; Life Care is globally distributed. |
| Energy Cost | High. Glass melting is extremely energy-intensive. Natural gas and electricity costs directly hit margins. | High. Same energy intensity. All-electric furnace technology (NEMT) is a partial mitigant. | Low. Mask blank production is precision-intensive, not energy-intensive. |
| Valuation Risk | Low. Trading below book value with 4%+ dividend yield. Limited downside from valuation compression. | Low-Medium. Trading near book value. Buyback provides floor support. | High. At 32x earnings, any execution miss triggers a disproportionate selloff. The stock fell from 29,000 to 14,345 within the past year -- a 50% drawdown. Premium stocks suffer premium drawdowns. |
| Key Risk Summary | Value trap. ROE stays at 5% forever, stock never re-rates. | Display cycle rolls over + yen appreciation = double headwind. | Semiconductor cycle downturn compresses the multiple from 32x to 20x = 37% drawdown. |
Bear Case Scenarios
| Scenario | AGC | NEG | Hoya |
|---|---|---|---|
| Bear Case Price | ~3,500-4,000 yen (0.5-0.6x book) | ~4,000-4,500 yen (below book, cycle trough) | ~14,000-15,000 yen (20-22x trough earnings) |
| Max Drawdown from Current | ~15-25% | ~28-36% | ~37-45% |
| What Triggers It | Global recession + energy spike + display downcycle | Severe display downcycle + yen appreciation to 120/USD | Semiconductor cycle bust + PENTAX restructuring failure + multiple compression |
The risk profiles are inverted relative to the return profiles. AGC has the least downside risk (trading below book with a fat dividend) but also the least upside. NEG has the most cyclical risk (operating leverage works in reverse). Hoya has the most valuation risk (paying 32x for perfection leaves no margin for error).
8. Technical Setup
| Technical Factor | AGC | NEG | Hoya |
|---|---|---|---|
| 1-Year Performance | Roughly +16% off 52-week low, but -35% from 52-week high | +108% from 52-week low (doubled), -13% from 52-week high | +65% from 52-week low, -20% from 52-week high |
| Trend (200-day MA) | Price near or below 200-day MA. Uptrend from April 2025 lows intact but weakening. | Price above 200-day MA (~5,500-5,700). Long-term uptrend intact despite recent 18% correction from 7,185 high. | Price above 200-day MA (~21,600). Uptrend intact with golden cross (50-day > 200-day). |
| Momentum (RSI) | Estimated 30-40 range (approaching oversold after sharp selloff from 6,959 to current levels) | Estimated 45-50 (neutral after partial recovery from March selloff) | RSI ~51 (neutral; stock has digested the rally) |
| Volume | Elevated volume on recent selloff suggests institutional distribution | Elevated selling volume during March 4-9 correction (2x average); normalized since | Below-average volume on recent sessions; rally may be losing participation |
| Support Levels | 4,000-4,200 (approximate 200-day MA + prior consolidation zone); 3,870 (52-week low) | 5,900-6,000 (March 9 low); 5,500-5,700 (200-day MA zone) | 22,500-23,000 (nearest support); 21,600 (200-day MA) |
| Resistance Levels | 5,500-5,800 (prior trading range); 6,959 (52-week high) | 6,500-6,700 (March consolidation); 7,185 (52-week high) | 29,175-29,590 (52-week high zone) |
| Chart Pattern | Pullback to rising 200-day MA in an uptrend. Classic buy-the-dip setup if the MA holds. | Bull flag / consolidation after strong rally. Holding above 200-day MA. | Consolidation / base-building in 23,000-29,000 range. Healthy digestion of 65% rally. |
| Technical Verdict | Moderately favorable entry. Approaching oversold near support. | Neutral. Long-term trend up, short-term messy. Wait or scale in. | Mixed-to-positive. Uptrend intact but extended. Better entry at 22,000-24,000. |
All three stocks sold off in early-to-mid March 2026, likely driven by broader market concerns (tariff fears, global growth worries). The correction has brought AGC and NEG closer to support levels, while Hoya has corrected less in percentage terms but still sits well below its 52-week high.
The best near-term technical entry is probably AGC, which is trading near its 200-day MA in oversold territory with a 4%+ dividend yield providing downside protection. NEG is second, consolidating after an 18% correction with the 200-day MA as a natural floor. Hoya is the worst technical entry right now -- still extended above its moving averages despite the correction, with 18% of air between the current price and the nearest meaningful support.
9. Composite Scorecard
Each dimension scored on a 1-5 scale. Weightings reflect what matters most for long-term total returns.
| Dimension (Weight) | AGC | NEG | Hoya |
|---|---|---|---|
| Business Quality (20%) | 2.5 | 3.0 | 5.0 |
| Financial Health (10%) | 4.0 | 4.5 | 5.0 |
| Growth (20%) | 2.0 | 2.5 | 4.5 |
| Valuation (15%) | 4.0 | 3.0 | 2.0 |
| Capital Allocation (10%) | 2.5 | 4.0 | 4.5 |
| Competitive Moat (15%) | 2.5 | 3.0 | 5.0 |
| Risk (5%) | 3.5 | 2.5 | 3.0 |
| Technical Setup (5%) | 3.5 | 3.0 | 2.5 |
| Weighted Total | 2.85 | 3.10 | 4.10 |
Score Justification
AGC (2.85/5): Cheap but low-quality. The valuation provides a margin of safety, but the business does not earn its cost of capital, growth has been nonexistent, and the conglomerate structure makes it hard to unlock value. The dividend yield is the best feature.
NEG (3.10/5): Middle of the road. The oligopoly position is genuine, the balance sheet is a fortress, and the buyback program is best-in-class for a Japanese mid-cap. But growth is low, ROIC barely covers WACC, and cyclical risk is real. The GC Core semiconductor optionality is a free call that could change the picture.
Hoya (4.10/5): Best business, worst price. The wide moat, consistent growth, world-class margins, and structural tailwinds make this the obvious long-term compounder. The only issue is valuation -- at 32x earnings, you are paying full price for perfection, which limits near-term upside and creates meaningful downside risk if the semiconductor cycle turns.
10. Final Verdict
Ranking
- Hoya (7741.T) -- The best business. By far.
- NEG (5214.T) -- The best risk/reward at current prices.
- AGC (5201.T) -- The cheapest, but cheap for a reason.
If You Can Only Buy One
Buy Hoya. Not because it is cheap -- it is not. Buy it because it is the only one of these three companies that consistently creates economic value above its cost of capital, has dominant positions in growing markets with no viable competitor, and has a management team that returns every yen of free cash flow to shareholders. Over a 5-year holding period, Hoya's combination of 7-9% revenue growth, 30%+ operating margins, and 3-5% annual share count reduction should compound EPS at 12-15% annually. At 28x forward earnings, that implies 12-15% annual returns from earnings growth alone, with additional upside if the multiple expands on High-NA EUV or HAMR catalysts.
The entry timing matters, though. At current prices (~23,700 yen), the stock is about 20% below its 52-week high and near Morningstar's 24,623 yen fair value estimate. This is a reasonable entry -- not a screaming bargain, but not overextended either. A scale-in approach makes sense: buy a starter position now (30-40% of target), add on any pullback to the 20,000-22,000 range (closer to the 200-day MA and below Morningstar fair value), and complete the position on either further weakness or a breakout above the 52-week high on volume.
If You Want the Best Near-Term Risk/Reward
Buy NEG. At 0.94x book value with a net cash balance sheet, 11.8% total shareholder yield (dividends + buybacks), and the display glass oligopoly providing structural protection, NEG offers the most attractive short-to-medium-term setup. The 100B yen buyback program puts a floor under the stock -- management is buying back over 5% of shares outstanding annually. If the display cycle holds and GC Core semiconductor glass materializes, there is meaningful upside to 7,500-8,500 yen. If the cycle turns, the book value floor (6,545 yen/share) and continued buybacks limit the downside.
NEG is the "get paid to wait" stock. You collect a 2.6% dividend, management buys back 9% of shares annually, and you have a free option on the semiconductor glass pivot. That is not a bad place to be.
If You Need Income Above All Else
Buy AGC. The 4.3% dividend yield is the highest of the three and is well-covered by free cash flow (96B yen FCF vs. ~45B yen dividend cost). The investment-grade balance sheet (S&P A-, Moody's A2) eliminates any dividend sustainability concern. AGC has held the 210 yen/share dividend through a year when it posted a 94B yen net loss -- that is the kind of dividend commitment income investors want to see.
But be honest about what you are buying. AGC is a 5% ROE conglomerate trading below book value for a reason. The dividend is the return. Capital appreciation is a bonus if the "AGC plus" portfolio transformation actually works, but you should not count on it.
The Diversification Play
If you want exposure to the entire Japanese glass substrate value chain and do not want to pick one, the allocation should be:
- 50-60% Hoya (the compounder, the anchor)
- 25-30% NEG (the cyclical upside, the buyback story)
- 10-20% AGC (the dividend ballast, the sum-of-parts option)
This gives you the best business (Hoya), the best near-term risk/reward (NEG), and the cheapest income stream (AGC). The three stocks have different return drivers -- Hoya is driven by semiconductor and healthcare megatrends, NEG by display cycle and buybacks, AGC by portfolio transformation and dividend -- so the correlation within the basket is lower than you might expect for three companies in the "same" industry.
What I Would Actually Do
I would put 60% in Hoya as a long-term compounder, 30% in NEG for the buyback and cyclical upside, and skip AGC entirely. AGC's 5% ROE is not good enough. The 4.3% dividend yield sounds attractive, but a company that earns below its cost of capital is slowly destroying value even as it pays you a dividend. You are getting your own money back, not earning a return on it. Life is too short to own value traps when you can own Hoya at a mid-range multiple.
The one thing that could change my mind on AGC: a meaningful increase in the buyback program. At 0.91x book value with 96B yen in free cash flow, AGC should be buying back 50-60B yen of stock per year. If management announced that, the stock would re-rate. Until they do, the passive capital allocation is the binding constraint on the investment thesis.
Report compiled March 17, 2026. Financial data sourced from company IR materials (AGC, NEG, Hoya), Yahoo Finance, StockAnalysis, Investing.com, MarketScreener, Morningstar, and prior research files (profiles, deep-dives, checklists). Forward estimates are labeled as management guidance, analyst consensus, or my estimates throughout. All yen figures in billions unless otherwise noted. Hoya's fiscal year ends March 31; AGC and NEG fiscal years end December 31.
Sources:
- Yahoo Finance -- AGC (5201.T)
- Yahoo Finance -- NEG (5214.T)
- Yahoo Finance -- Hoya (7741.T)
- StockAnalysis -- Hoya (TYO:7741)
- StockAnalysis -- AGC (TYO:5201)
- StockAnalysis -- NEG (TYO:5214)
- MarketScreener -- AGC Consensus
- MarketScreener -- Hoya Consensus
- Investing.com -- AGC
- Investing.com -- NEG
- Investing.com -- Hoya
- Morningstar -- AGC (5201)
- Morningstar -- Hoya (7741)