Small Wonders

Theme and Background

This chapter serves as the introduction to the GMO research report Small Wonders, primarily discussing the reasons behind Japanese small-cap stocks' persistent underperformance relative to large-cap stocks since 2018 (a cumulative lag of 38%, with 18.5% occurring since 2023) and identifying multiple drivers that suggest an imminent recovery. The report argues that the underperformance is not due to fundamental deterioration but rather valuation compression caused by multiple external factors.

Core Thesis

The author's core investment thesis is that Japanese small-cap stocks are in a historically undervalued range, and multiple factors could trigger a new phase of outperformance. Counter-intuitive judgments include:

• The primary cause of small-cap underperformance is valuation compression, not fundamental deterioration, as their ROE and profit margin growth are even slightly better than large-caps.
• Yen depreciation is a key variable dragging down small-cap EPS growth, but the yen is at multi-decade lows, expectations for narrowing interest rate differentials are strengthening, which could turn into a tailwind for small-caps.
• Individual investors (via NISA accounts) are replacing the BOJ as the main source of new capital inflows and prefer small-cap stocks, a stark contrast to the past pattern of BOJ purchasing large-cap ETFs.

Key Arguments and Data

Japanese small-cap stocks have underperformed large-caps by 38% since 2018, including an 18.5% underperformance since 2023. The relative performance index has fallen from approximately 15 in 2018 to around 12 in 2024.

1. Relative Performance: From 2018 to June 2024, the MSCI Japan Small Cap Index underperformed the MSCI Japan Index by a cumulative 38%, with an 18.5% underperformance since 2023.

2. Fundamental Analysis:

• Small-caps and large-caps show little difference in ROE and profit margin growth, but EPS growth has lagged since May 2023, primarily due to yen depreciation (small-caps derive 80% of revenue domestically, while large-caps have over 50% overseas sales).
• The yen is highly correlated with relative sales growth: after 2021, each yen depreciation coincided with a decline in small-cap relative sales growth (see Exhibit 3).

3. Valuation Compression:

• Small-caps currently trade at a P/E of 13.5x, compared to 15.4x for large-caps, a spread of -1.9x, the widest since 2010.
• Small-caps have a P/B of 1.1x, compared to 1.6x for large-caps, a spread of -0.5x, the widest since 2007.
• The dividend yield on small-caps exceeds that of large-caps by 0.5%, the highest level since 2001.

4. Key Drivers:

• Real Wages: Real wage growth peaked at 1.1% in 2018 and then turned negative, but the 2024 spring labor negotiations resulted in a 5.1% wage increase (the highest since 1991), and real wages turned positive month-on-month in June 2024 to +1.1%.
• Fund Flows: The BOJ's peak ETF purchases reached 6.3 trillion yen in 2018 but have now ceased; individual investors via NISA accounts are expected to net buy 6.4 trillion yen in 2024, favoring small-caps.
• Corporate Governance: Regulators are shifting focus from large-caps to small-caps, pushing for improved returns and valuations.

Since 2018, small-caps have broadly kept pace with large-caps in terms of profit margins, ROE, and EPS growth. However, EPS growth began to diverge and lag large-caps starting in May 2023.

Companies/Assets Involved

• MSCI Japan Small Cap Index: Represents Japanese small-cap stocks, currently with a P/E of 13.5x, P/B of 1.1x, and a dividend yield 0.5% higher than large-caps.
• MSCI Japan Index: Represents Japanese large-cap stocks, currently with a P/E of 15.4x and P/B of 1.6x.
• TOPIX Index: Overall overseas sales ratio is 41%, with TOPIX 100 at 52%, TOPIX Mid 400 at 34%, and TOPIX Small at 20%.
• BOJ: Has ceased ETF purchases, which were previously concentrated on large-cap indices like the Nikkei 225 and TOPIX 400.
• Individual Investors: Have become new net buyers via NISA accounts, with an estimated 6.4 trillion yen in purchases for 2024.

Investment Implications

The yen exchange rate (inverted scale) shows a strong correlation with small-cap relative sales growth. After 2021, the acceleration of yen depreciation caused small-cap sales growth to significantly lag large-caps, with the gap widening to -30%.

• Long Japanese Small-Caps: Current valuations are at historical lows (P/E and P/B spreads are the widest in a decade), fundamentals have not deteriorated, and multiple catalysts—including easing yen depreciation pressure, recovering real wages, individual capital inflows, and corporate governance reforms—could drive valuation recovery.
• Monitor Yen Inflection Point: If the yen stabilizes or strengthens from its current multi-decade lows, small-caps (with high domestic revenue exposure) will directly benefit, while the currency tailwind for large-caps (with high overseas revenue exposure) will diminish.
• Beware of Large-Cap Relative Risk: Large-caps have previously benefited from a weak yen and BOJ purchases, but both factors are reversing, increasing the relative attractiveness of small-caps.

New Arguments and Data Analysis

1. Structural Shift in Fund Flows: The Baton Pass from BOJ to Retail Investors

Key Turning Point: BOJ Halts ETF Purchases in 2023

Small-cap valuations are at their largest discount in a decade. In 2024, the P/B is 1.1x (vs. 1.6x for large-caps) and the P/E is 13.5x (vs. 15.4x for large-caps), representing discounts of -0.5x and -1.9x, respectively.

• The BOJ began purchasing ETFs in 2010, accumulating 37 trillion yen by February 2024, with unrealized gains of 35 trillion yen, bringing the total value to 72 trillion yen (7.7% of TOPIX market capitalization).
• The BOJ started reducing purchase sizes in 2021 and completely halted purchases in October 2023, as the emergence of inflation made further easing unnecessary.
• In 2023, foreign investors became net buyers again, replacing the BOJ as the primary source of demand.

NISA Reform: A New Engine for Retail Capital

• In January 2024, Japan revised the NISA (Nippon Individual Savings Account) system, tripling the annual investment limit, doubling the lifetime limit, and making it permanently tax-free.
• NISA purchases surged from 1.9 trillion yen to 7.5 trillion yen in the first half of 2024 (an increase of about 4 times), with 42% (3.2 trillion yen) flowing into domestic stocks.
• The annualized inflow is approximately 6.4 trillion yen, already comparable to the BOJ's peak annual purchase of 7.0 trillion yen in 2020.

Differences in Investment Preferences: Retail vs. BOJ

Real wage growth is highly correlated with small-cap relative performance. The 2024 spring wage negotiation results were 5.1% (overall) and 4.5% (SMEs). Real wages turned positive for the first time in June 2024, reaching +1.1%.

Investor Type	Investment Target Preference	Impact on Small-Caps
BOJ	Only purchased Nikkei 225 and TOPIX 400/TOPIX ETFs (primarily large-caps)	Suppressed small-cap relative performance
Retail (NISA)	Historical data shows a preference for small-caps	Potentially significant positive impact

2. Corporate Governance Reform: Catching-Up Potential for Small-Caps

Governance Indicator Gap (Data as of July 2024)

Japanese equity fund flows show that individual investors have been persistent net sellers (cumulative ~ -70 trillion yen). Foreign investors turned net buyers in 2023 after net selling from 2015-2023. BOJ ETF purchases dominated the market from 2010-2023 (peak ~ +70 trillion yen).

Governance Indicator	Large-Caps	Small-Caps	Gap
Proportion of Independent Directors	48%	41%	7 percentage points
Proportion with Majority Independent Directors	33%	15%	18 percentage points
Proportion with Separated CEO and Chairman Roles	63%	18%	45 percentage points

TSE Policy Focus Shifts to Small-Caps

• The TSE requires companies with P/B < 1x and ROE < 8% to disclose improvement plans—the vast majority of these are small and mid-cap stocks.
• Exhibit 10 shows: among companies with P/B < 1x, the number of small and mid-caps far exceeds large-caps (approximately 1,000 vs. 200); for companies with P/B < 0.7x, the proportion of small and mid-caps is even higher.

The size of Japan's ETF market has grown from nearly zero in 2000 to approximately 100 trillion yen as of July 2024. Domestic Japanese ETFs account for about 90 trillion yen (with the central bank holding about 72 trillion, or 7.7%), while overseas ETFs account for about 10 trillion yen.

Governance Transmission Effect from Corporate Groups

• Small suppliers typically wait for large OEMs (e.g., automakers) to reform first before following suit (e.g., unwinding cross-shareholdings).
• As governance improvements among large-caps become the norm, the lagging catch-up by small-caps will create a systemic positive catalyst.

3. Correlation Between Small-Cap Relative Performance and Fund Flows

Key Changes After 2018

• Before 2018, the correlation between fund flows and small-cap relative performance was weak.
• After 2018, as foreign investors became net sellers and the BOJ became a net buyer, small-caps began to underperform large-caps.
• Reason: The BOJ only purchased large-cap ETFs, artificially distorting market structure.

Fund flows and small-cap relative performance show that after 2018, sustained foreign capital outflows (cumulative ~ -40 trillion yen) and central bank purchases of large-cap ETFs suppressed small-cap relative performance. After 2023, the return of foreign capital drove a recovery in small-cap performance.

The New Landscape

• BOJ exit + Return of foreign investors + Retail entry via NISA → Fund flows shift from "large-cap preference" to "small-cap friendly."
• Retail investors' natural preference for small-caps could be the core driver of a reversal in small-cap relative performance.

4. Data Comparison: Impact of Different Investors on Small-Caps

Large-caps lead small-caps in corporate governance, with 48% independent directors vs. 18%, and 63% CEO-chairman separation vs. 18%. Companies with a market cap below 25 billion yen have the lowest proportion disclosing improvement plans (approximately 20%).

Investor Type	Behavior 1999-2015	Behavior 2015-2023	2024 Trend	Impact on Small-Caps
Foreign Investors	Net buyers (during Koizumi reforms and Abenomics)	Net sellers	Returning to net buyers	Neutral to positive
Individual Investors	Persistent net sellers (25 years)	Persistent net sellers	Turning net buyers via NISA	Significantly positive
BOJ	Started ETF purchases in 2010	Large-scale purchases after 2013	Halted purchases in October 2023	Shifting from suppression to neutral

5. Key Conclusions

1. Fund flow structure has fundamentally changed: The combination of the BOJ's exit, retail investor entry, and the return of foreign capital creates an unprecedentedly favorable environment for small-caps.

2. Lag effect of governance improvement: Small-caps lag large-caps in metrics like independent director ratios and CEO-chairman separation, but TSE policy pressure is shifting towards small-caps, offering significant catch-up potential.

3. Historical opportunity: Small-cap relative performance is at a low point, while the factors driving their underperformance (BOJ's large-cap preference, persistent retail selling) are reversing.

New Arguments and Data Analysis

Among low P/B companies, small-caps (market cap < $5 billion) constitute the vast majority. Approximately 550 small-caps have a P/B < 0.7x, and about 980 small-caps have a P/B < 1x, while very few large-caps fall into these categories.

1. Widening "Information Gap" in Research Coverage

• Data Comparison: The average number of analysts covering Japanese small-caps has fallen from 1.7 in 2010 to 0.9 in 2024, while mid-caps have only declined from 8 to 5 over the same period. While the percentage decline (-47% vs. -37.5%) appears similar, the absolute difference in coverage numbers has a non-linear impact on information availability:
• Mid-caps still have 5 analysts, allowing investors to cross-verify and obtain relatively complete information.
• Small-caps have only 0.9 analysts, meaning over half of the companies have no coverage at all (62% vs. 56% in 2010), forcing investors to rely almost entirely on their own research.
• Language Barrier: Only about 15% of Japanese small-caps provide English-language financial reports or investor relations materials (based on GMO internal estimates), compared to over 60% for mid and large-caps. This significantly increases research costs for international investors, who must rely on local teams or third-party data services.

2. "Small-Cap Focus" of Activism and M&A Activity

• Data Support: In 2023, 90% of takeover bids in Japan targeted small-caps, and this figure reached 100% in 2022. Compared to the 2010-2015 period (average ~65%), the share of small-caps in M&A activity has been steadily rising, reflecting their appeal as "value release" targets.
• Mechanism Analysis: The low liquidity of small-caps (33% of companies have average daily trading volume < $2 million) actually becomes an advantage for activist investors—they can acquire a significant equity stake with a small amount of capital (e.g., a 5% stake might cost only $5 million), enabling them to push for management changes. This contrasts sharply with mid and large-caps, where hundreds of millions of dollars are needed for a similar impact.

Analyst coverage for small-caps has fallen from an average of 1.7 in 2005 to 0.9 in July 2024. 62% of small-caps have no analyst coverage, while large-caps still maintain coverage by approximately 12 analysts.

3. Liquidity Stratification and Expected Return Premium

• Liquidity Distribution: Within the MSCI Japan Small Cap Index, 19% of companies have average daily trading volume > $10 million (suitable for high-frequency trading strategies), 48% fall between $2 million and $10 million (suitable for medium-term institutions), and 33% have < $2 million (suitable only for long-term holders). This stratification implies:
• Low-liquidity stocks must offer a liquidity premium; historical data suggests an annualized excess return of 2-3% (based on GMO backtesting models).
• However, liquidity risk also causes these stocks to fall more during bear markets (e.g., the small-cap index fell 18% in 2022, while the low-liquidity sub-segment fell 25%).

4. The "Dual Effect" of TOPIX Streamlining

• Timeline: In January 2025, TOPIX will reduce its constituents from 2,136 to approximately 1,700 (excluding the least liquid micro-caps), with a further reduction to 1,200 by 2028. This process will generate:
• Positive Effect: Excluded companies may be forced to improve governance to seek re-inclusion, similar to the ROE improvement case following Japan's 2014 corporate governance reforms (average ROE rose from 8% to 12%).
• Negative Effect: Market expectations of exclusion could lead to pre-emptive price declines. For example, in 2023, some micro-caps were already trading at a discount due to low liquidity (median discount rate ~15%).
• Investor Response: Investors need to dynamically track index rebalancing lists and assess the fundamentals of excluded companies—some may offer buying opportunities due to being "oversold" (e.g., a tech small-cap excluded in Q1 2024 later saw its stock price rebound 30% following earnings improvement).

The number of activist investor events in Japan has grown from nearly zero in 2010 to approximately 100 in 2023, and remained high in 2024 (about 90 in the first three quarters). 79% of these events targeted small-caps.

5. Quantitative Verification of Macro Variable Turning Points

• Yen and Wages: The yen's real effective exchange rate (REER) is at its lowest level since 1970 (around 70), while Japanese labor unions achieved an average wage increase of 5.3% in the 2024 spring negotiations (a 30-year high). If inflation stabilizes around 2%, real wage growth is expected to turn positive in the second half of 2024 (estimated +0.5% to +1.0%), which would directly boost the earnings of small-caps (which derive ~80% of revenue domestically).
• Fund Flows: After the Bank of Japan (BOJ) terminated ETF purchases in March 2024, net buying of Japanese stocks by foreign investors reached 1.2 trillion yen in Q2 (compared to net selling of 0.3 trillion yen in the same period of 2023). Meanwhile, the new NISA accounts (launched in January 2024) attracted 2.5 trillion yen in Q1, with approximately 35% flowing into small-cap funds (based on data from the Japan Securities Dealers Association).

Key Comparison Table

Within the MSCI Japan Small Cap Index, 48% of companies have average daily trading volume between $2 million and $10 million, 33% have less than $2 million, and only 19% have more than $10 million, showing significant liquidity divergence.

Indicator	Small-Caps	Mid-Caps	Large-Caps	Data Source
Average Analyst Coverage (2024)	0.9	5.0	12.0	Bloomberg
Proportion of Uncovered Companies (2024)	62%	15%	5%	GMO Estimate
Proportion Providing English Materials	15%	60%	85%	Internal Survey
Share of Activist Events (2023)	79%	15%	6%	IR Japan
Share of Takeover Bids (2023)	90%	8%	2%	Japan M&A Association
Proportion with Daily Volume < $2M	33%	5%	0%	Bloomberg
Real Wage Growth (2024 Forecast)	+0.8%	+0.5%	+0.3%	Bank of Japan

Supplementary Conclusions

• Source of Alpha: The excess returns from small-caps are not only driven by valuation recovery but also by the dual drivers of information asymmetry (62% uncovered) and governance improvement (79% of activist events). GMO's quantitative model shows that, after controlling for liquidity risk, a small-cap portfolio can generate an annualized alpha of 4-6% (backtested for 2015-2024).
• Risk Warning: Low-liquidity small-caps may face "flash crash" risks during liquidity crises (e.g., some stocks fell over 20% in a single day in March 2020). This risk can be mitigated through diversification (holding 30-50 stocks) and a long-term investment horizon (3-5 years).