AFM State of the Capital Markets 2025

State of the Capital Markets 2025
In short Please find this year’s AFM State of the Capital Markets. Investors make investment decisions based on expected future returns. Economic developments, the level of interest rates and future expectations are still important factors that determine price formation. However, there are other factors that influence the price formation process in today's capital markets, especially in the short term. Sudden changes in sentiment due to political developments, price developments of correlated instruments and directional messages on social media are just a few examples of elements that can have a direct price impact. This version of the State of the Capital Markets contains a number of images that provide insights into the price formation process based on current data. ANALYSIS REPORT

AFM State of the Capital Markets | June 2025 Asset class: Equity
Figure 1. Time ‘at the touch’ top 1, 3 and 5 firms
Asset class: Equity
Figure 2. Concentration and spread
What does the data show?
This chart displays the market share at the touch for the top 1, 3, and 5 trading firms, ranging from the least traded shares to the most actively traded ones. What does it imply? On average, the top firm is at the touch 31% of the time, the top three firms 58% of the time, and the top five firms 73% of the time. Generally, as trading activity increases, market share concentration at the touch decreases across all three measures. However, this trend reverses for the most heavily traded shares, which exhibit increased concentration among the top firms. On average, the top 3 firms are at the touch 58% of the time. Source: Euronext, shares in Dec 2024 What does the data show? The concentration of firms "at the touch“ in shares — that is, those oering the best bid or best oer—is positively related to the spread of a stock. What does it imply? Intuitively, competition among firms encourages them to outbid each other on the buy side or undercut each other on the sell side, leading to a narrower spread (the dierence between the best oer and best bid prices). Therefore, the more firms participate in this process, the tighter the spread tends to be. 20 40 60 80 100% 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 Spread in basis points (log10) % time at-the-touch top 3 firms Source: Euronext, shares in Dec 2024 ANALYSIS REPORT 1 Top firm Top 3 firm Top 5 firm % time at-the-touch Least Most traded Deciles of shares Average How to read this graph? 0 20 40 60 80 100 The more competition to provide liquidity, the lower this % will be - and vice versa.

What does the data show?
This chart shows the median value at the top level of the order book throughout the trading day for shares in the three Dutch indices: AEX, AMX, and ASCX. What does it imply? The chart shows that large-cap shares tend to have significantly higher values at the top of the order book compared to smaller-cap shares. For instance, AEX shares typically show values around four times higher than AMX shares, and five times higher than ASCX shares. Within the AEX index, there is a clear upward trend in order book value throughout the day, with a slight dip in the early afternoon. This upward momentum resumes after the U.S. markets open. 0 10,000 20,000 30,000 40,000 50,000 Source: Euronext, shares in December 2024 What does the data show? This chart shows how the trading in shares has been internationalized in the current market landscape. The trading firms are clustered per country. Reported retail transactions have been categorized separately. What does it imply? Most trading in Dutch shares is by non-Dutch entities. British firms carry out the majority of transactions on the AEX, AMX, and ASCX indices. On average only 17% of the trading in AEX share are by Dutch trading firms. For regulatory obligations, licensed trading firms have to fulfil their obligations with the NCA in which they applied for the license. This makes international cooperation between European regulators essential. Source: Euronext, shares in December 2024 Trading shares: Top 5 AEX United Kingdom 24% France 21% Ireland 20% Netherlands 17% Cayman Islands 5% Other Trading shares: Top 5 AMX United Kingdom 34% Ireland 18% France 17% Netherlands 9% Retail 6% Other Trading shares: Top 5 ASCX United Kingdom 32% Retail 16% Netherlands 14% Ireland 13% France 13% Other Asset class: Equity
Figure 3. Value ‘at the touch’ during trading
Asset class: Equity
Figure 4. Nationality counterparties Dutch shares
AEX
AMX
ASCX
Value at-the-touch (€)
ANALYSIS REPORT
AFM State of the Capital Markets | June 2025 2

What does the data show?
This figure shows the percentage of traded value during the Open, Regular Trading Hours, and Closing Auction on Euronext Amsterdam. We observe an upwards trend in the share of trading taking place during the Closing Auction, with the percentage in the Closing Auction in 2025 (until now) being almost 10 percentage points higher than in 2020. What does it imply? This represents a significant rise over just the past four years. It is particularly interesting because the Closing Auction is where genuine supply and demand meet the market, whereas during regular trading hours, volumes are also partially driven by day trading and speculation, which do not necessarily reflect true underlying demand and supply. Source: Euronext Amsterdam shares ADYEN, ASML, SHELL, PHILIPS 0 20 40 60 80 % of the value traded 2020 2025 Open Trading Close Euronext Amsterdam:
A significant increase in trading share during Closing Auction. It is here where institutional supply and demand meet the market. What does the data show? Despite the increase of volume in the Closing Auction, the dierence between the last traded price and the auction price has been stable over the last few years. This chart is a sample of the share ING, which is a representative stock in the AEX index. What does it imply? Not only in the Netherlands, but also in other EEA jurisdictions, trading moves more and more to the close of the trading day. Besides the increase in volume of the closing auction, also the last trading hours of the day attracts more investors, also because American markets are open as from 15.30 CET. During the 5 minute auction period, large orders do have a temporary impact on the price, but eventually these large orders seem to be absorbed by the market, leaving a relative small impact on the auction price. Source: MiFIR data, share ING 0 5 10 15 20 25 Dierence between last and auction price (in €0,01) 2022 2023 2024 2025 The dierence between the last traded price and the auction price has been stable. Asset class: Equity
Figure 5. Percentage value traded in the Close
increases
Asset class: Equity
Figure 6. Dierence between last and auction
price
ANALYSIS REPORT
AFM State of the Capital Markets | June 2025 3

What does the data show?
This graph shows the venues where retail orders are executed. This image is a sample of one month of trading (January 2025) in one of the major bluechip stocks. Out of 85 venues included in this sample, 38 showed actual trading activity. The largest 4 venues were responsible for 90% of the execution of retail orders. Where almost half of the retail orders were executed on one RM, and almost 40% of the orders were executed on (foreign) MTF’s. What does it imply? A relatively small part of the retail orders were executed via a Systematic Internaliser. For trading firms and -venues, so-called uninformed retail orders are attractive. This leads to competition between venues to attract these stream of orders, often in alliance with the commitment of trading firms. Each dot is a venue (85 in total) during one month of trading. The colours respond to trading activity:
<45%
10-20%
<2%
No activity
The largest 4 venues were responsible for 90% of the execution of retail order Regulated market (RM) Multilateral Trading Facility (MTF) Systematic Internaliser (SI) Asset class: Equity
Figure 7. Execution Retail Orders
ANALYSIS REPORT
AFM State of the Capital Markets | June 2025 4 Source: MiFIR data d e What does the data show? We express intraday price movements as the ratio between intraday closing price changes and the intraday price movements. Accumulated intraday price movements are the result of comparing every trading price with the previous trading price and summing the absolute values of these price movements. Here we compare these intraday price movements with interday closing price changes. This ratio falls between 0 and 1 (absolute summed intraday changes are one directional and equal to the interday change). What does it imply? The lower the ratio, the more two-way trading takes place in relation to the actual direction of the share price. This means that market participants have more possibilities to absorb the actual price movement in their trading activities (i.e., ‘liquidity’). The graph shows that the liquidity in the AEX is superior to AMX and AScX, but has decreased over the last five years. Asset class: Equity
Figure 8. Intraday price movement decreases
in relation to interday change
0,00
0,05
0,10
0,15
0,20
0,25
‘21 ‘22 ‘23 ‘24 ‘25
Ratio between intraday closing price changes and accumulated intraday absolute price movements AScX ‘21 ‘22 ‘23 ‘24 ‘25 AMX ‘21 ‘22 ‘23 ‘24 ‘25 AEX For AEX, we see a persistent increase in the ratio in the past few years, indicating a decrease in liquidity. a b c a Maximum b 3rd quartile c Median d 1st quartile e Minimum Source: MiFIR data

0
200
400
600
800
1,000
Sum of normalised AEX (ETF)
Sum of AEX Index - Last Price (L1)
29 Apr 2020 26 Feb 2025
What does the data show?
ETFs on indices are popular investment instruments which equal the performance of an index with the benefit of the tradability of shares. Especially in the US, ETFs are very popular with retail investors. The ETFs on the S&P 500 are among the most liquid derivative instruments in the world. The supply and demand in those ETFs at times even determines the price of the underlying shares of the index. What does it imply? For longterm investors, correlation between ETFs and indices turns out to be nearly 100%, and this explains the popularity of the instrument under investors. A correlation of almost 100% Index Source: Bloomberg Asset class: Derivatives
Figure 10. Correlation AEX vs AEX ETFs
long term
ANALYSIS REPORT
What does the data show?
This graph shows the rolling 10 day average dierence in points between the Open and Close of the AEX and the (overnight) Close and Open. The timeframe for this graph is one month, from half March to half April
2025. Given the market turbulence in the beginning of April due to the
announced import taris, most of the significant pricemoving factors originated in the US. What does it imply? One could say that the priceforming process was highly impacted by the geopolitical developments in the beginning of April. This led to large price moves outside the European trading hours. 10 20 13 April 2025 14 March 2025 Average dierence in points The US announced import taris, resulting in price moves overnight. Close > Open (overnight) Open > Close (day time) Asset class: Derivatives
Figure 9. AEX Dierence Open - Close - Open
Source: MiFIR data
AFM State of the Capital Markets | June 2025 5

Source: Bloomberg
871
872
873
874
875
876
877
878
879
AEX
ETF Normalized
Traded price in €
On an average trading day
09:00:30 15:24:15
What does the data show?
During an average trading day, the traded price of an ETF may deviate from the value of the correlated underlying index. This deviation is irregular, but the trend line does correspond to the index price. What does it imply? The price of the ETFs is primarily driven by retailorders, which interact with liquidity providers. The value of the index is frequently calculated based on the price of the index shares. The ETFs only result in a datapoint when there is a transaction. Asset class: Derivatives
Figure 11. Correlation AEX vs AEX ETFs
short term
ANALYSIS REPORT
AFM State of the Capital Markets | June 2025 6

ANALYSIS REPORT
Asset class: Commodities
Figure 12. Dutch TTF gas and CO2 emission
prices over time
Asset class: Commodities
Figure 13. Correlation between TTF and CO2
prices
160
140
120
100
80
60
40
20
0
Jan 2024 Feb 2025
Dutch TTF gas price
CO2
Emissions Futures price
What does the data show?
This graph shows the evolution of Dutch TTF gas prices and EU CO₂ emission allowance (EUA) settlement prices over time. What does it imply? The data implies that both gas and carbon settlement prices exhibit fluctuations, but they do not always move in tandem. Periods of divergence suggest diering drivers behind gas market and carbon market dynamics. Explanation The TTF gas settlement price reflects supply and demand dynamics in the European gas market, influenced by factors like weather, storage levels, and geopolitical events. The EUA settlement price represents the cost of emitting one tonne of CO₂ and is driven by EU climate policy and industrial demand. 90 120 150 20 30 40 50 60 TTF price CO2 price (€) Start: 2 Jan 2024 End: 19 Feb 2025 While energy prices and carbon costs can be related (e.g., through fuel switching or economic activity), this graph suggests that their price trends are not perfectly aligned, indicating a more complex relationship. AFM State of the Capital Markets | June 2025 7 Source: MiFIR data Source: MiFIR data Price

Asset class: Commodities
Figure 14. Investment funds position in
2024 -TTF Gas Futures
What does the data show?
This chart shows the net (i.e., Long minus Short) derivative positions in TTF Gas Future Contracts traded on ICE Endex, categorized by sector. These positions are calculated using the monthly average of daily aggregated position. The unit of measurement is MWh, adjusted by the relevant price multiplier. A large positive number indicates a dominance of long positions within the respective sector group. What does it imply? A clear directional trend in investment funds' trading strategies becomes apparent in TTF futures starting from late March 2025. The maximum net position is reached at the end of November, just below 290 million MWh. These long positions are balanced by short positions held by banks, investment firms, and commercial entities. Banks / investment firms Commercial entities Funds Net positions, millions MWh Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 2024 -300 300 -200 200 -100 100 0 A clear increasing trend in investment funds' trading strategies Asset class: Commodities
Figure 15. Hedging strategies positions in
2024 - TTF Gas Futures
The directionality of the two trading strategies increased in 2024. -200 -100 100 200 300 0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 2024 Net positions, millions MWh Funds Hedging strategies What does the data show? This chart shows the net (i.e., Long minus Short) derivative positions in TTF Gas Future Contracts traded on ICE Endex. Positions are classified as either held by Investment Funds (speculative) or by Commercial Entities hedging their physical spot market exposure (hedging strategies). A large positive number indicates a dominance of long positions over short positions. However, a large positive or negative value does not necessarily imply greater market participation—rather, it reflects an imbalance in positioning toward long or short. What does it imply? This chart is a subset of the previous one, focusing on two specific trading strategies. This figure shows that directional strategies in hedging can be equally important when evaluating market trends as the directional strategies from funds. Source: MiFIR data Source: MiFIR data ANALYSIS REPORT AFM State of the Capital Markets | June 2025 8

ANALYSIS REPORT
What does the data show?
This chart shows the median repo rate for centrally cleared and bilateral transactions over time. In cleared repo transactions, a central counterparty inserts itself between two counterparties and becomes the new counterparty in both legs of the transaction (novation). Some of the negative spikes may be due to issues with data quality. On year end, and to a lesser extent on quarter ends, negative spikes in rates are due to balance sheet scarcity of dealers around reporting dates, resulting in temporary lower market intermediation. What does it imply? Firstly, the repo rate is mainly influenced by the central bank policy rate (for the deposit and lending facilities), which acts as an anchor point. Multiple rate reductions by the central bank are therefore also visible in the repo rate. Secondly, the repo rate for cleared transactions is not much dierent than for bilateral transactions. Costs and benefits of centrally clearing repo transactions (netting exposures, exchanging margin etc.) do not seem to lead to a large spread in the repo rates between both transaction types. Source: SFTR (AFM calculations) 2,5 3,0 3,5 4,0 Repo rate, % Jan 2024 Jan 2025 Cleared Bilateral On year end, negative spikes are due to balance sheet scarcity of dealers around reporting dates, resulting in temporary lower market intermediation. In 2024, the central bank reduced rates multiple times. This is visible in the repo rates. Asset class: Fixed income
Figure 16. Repo rate of cleared vs bilateral
transactions
AFM State of the Capital Markets | June 2025 9 What does the data show? The Dutch State Treasury Agency is responsible for the placement of the Dutch State Loans. For this purpose, there are 13 Primary Dealers selected who have quoting obligations. This gives the Dealers the opportunity to be in competition to take up the DSL’s in the primary market. This graph shows that not all 13 PD’s are in competion in the primary market. There is a significant dierence between the top 6 and the other Dealers. What does it imply? The Dutch State Loan Agency uses a duration weighed system to rank the Primary Dealers. The goal is to trigger the PD’s to cater for a liquid market and to create a balanced supply and demand for the diverse durations. This ranking explains the dierences in the stake of the several PD’s as shown in the graph above. Asset class: Fixed income
Figure 17. Placement of state loans with
Primary Dealers
0 20 40 60 80 100
The top 6 PD’s make a total of circa 80% of the total (13 PD’s). Dealer 1-6 Dealer 7-13 % of placement Primary Dealers (PD’s) are banks appointed to purchase, promote and distribute Dutch government bonds. Source: MiFIR data