Finance

The latest reporting from the Financial Times highlights a point that energy analysts have been making for years: geopolitical shocks consistently strengthen the case for renewables, electrification and storage. Microsoft’s global vice-president for energy notes that oil and gas price spikes linked to the Middle East conflict reinforce the value of wind, solar and batteries in providing price stability. Once installed, renewables offer predictable cost profiles and reduce exposure to volatile global fuel markets. We saw this dynamic after Russia’s invasion of Ukraine. Europe accelerated solar deployment, heat pump uptake increased in several countries, and governments revisited questions of energy security through the lens of diversification and electrification. The underlying issue remains unchanged. Fossil fuels must continuously flow through complex global supply chains. When those flows are disrupted, prices spike and economies are exposed. Renewables, by contrast, are capital intensive upfront but deliver long term domestic supply and insulation from commodity shocks. There are short term risks. Inflation, higher interest rates and supply chain constraints can slow clean energy investment. Some governments may also respond by doubling down on gas infrastructure. The policy challenge is to avoid locking in further structural vulnerability. Energy security and climate policy are not competing objectives. In a world of recurrent geopolitical instability, they are increasingly aligned.

Energy is once again dominating headlines all over the world. Gas and oil prices are volatile, key shipping routes face geopolitical pressure, and policymakers are concerned about supply risks. The renewed uncertainty is a reminder of an uncomfortable reality: the next energy crisis isn’t an if – it’s a when, and a question of how prepared we are. A defining challenge of this decade, and one that now feels more urgent than ever, is how to build a resilient energy system. One that minimises structural dependencies and is designed for rising electricity demand. The imperative of our time: The more we electrify, the less we import fossil fuels. The less we import, the more resilient we become. The course of action is clear: ▪️ Relentlessly scale renewables: Slowing the buildout will not reduce costs. Quite the opposite – delay compounds system costs for the entire economy. ▪️ Fix the grids: As fast as possible, as efficiently as possible, and at the lowest possible cost. Before they become even more of a bottleneck. ▪️ Secure 24/7 electricity supply: When the wind isn’t blowing and the sun isn’t shining, renewables need reliable backup in the form of battery storage and hydrogen-ready gas fired power plants. But gas should serve only as a backup, with renewables and batteries reducing its utilisation. ▪️ Reduce gas supply dependence with infrastructure and diversification: We must not replace old dependencies with new ones. Diversification of gas supplies is key. And the physical prerequisite is an import infrastructure with buffers. We need the planned LNG terminals, complemented by a nationally held gas reserve to help ensure secure supply in winter. ▪️ Electrify everything that makes sense: The more we can power with mostly homegrown electrons, the less dependent we become on fossil imports. Other energy import-dependent countries like Japan and China have electrification rates that are around 10 percentage points higher than Germany’s. This shows where the path forward lies. Electrification reduces reliance on imported fossil fuels, which in turn strengthens overall resilience. The time to act is now.

Every business, small or large, private or publicly listed has to choose between borrowed money (debt) and owner funds (equity), and in this sessions I start by looking at the fictional reasons (debt is cheaper than equity, debt increases ROE), the real reasons (the tax benefits of debt vs bankruptcy costs) and the frictional reasons (desire for control, subsidized debt, and protections against bankruptcy). I then look at tax rates (marginal and effective) in 2025 as well as developments on the default front (ratings changes, loan defaults) during the year, before chronicling what companies around the world looked like both on debt comfort ratios (interest coverage and debt to EBITDA) and debt loads (debt to capital). I close by looking at two developments - the immense cap ex in AI and the growth of private credit, and argue that there is a big market delusion embedded here, and when it corrects, it will create a clean up and shrinkage in both.

The importance of a Cofounder Agreement Back in April 2017, I started my first startup with a close friend from college. We had known each other for about 4 years and started off on a positive note, incorporating a company with an equal 50/50 split. Soon after, we got incubated at SINE, IIT Bombay’s startup incubator. As part of the process, SINE made it mandatory for us to sign a Cofounder Agreement. At that time, we were just a few months out of college and didn’t fully appreciate its value. We googled a format, customised it, and signed. One of the clauses was about a lock-in period: "The Founders hereby agree that the shares held by them in the Company shall be locked in for a period of [] years ("Lock-in Period") from the Execution Date..." We decided to put 2 years as the lock-in period, without giving it much thought. This agreement was signed on 29th July 2017. Fast forward to 6th August 2018, barely a year later, my cofounder quit. At that time, he owned ~50% of the company. The only reason I could save the company was because of that lock-in clause. Without it, half the company would have gone to someone who had already exited. That experience taught me a few lessons: 1. Legal agreements can be lifesavers when things go wrong. 2. Not everyone thinks long-term. 3. People can quit abruptly without notice. Since then, I’ve always been very particular about legal agreements, especially termination and lock-in clauses. They protect not just you, but also your team, customers, and investors. If you’re running a startup and haven’t signed a Cofounder Agreement yet, please do it now. It’s one of the best favours you can do for your future self. #startups #business #entrepreneurship

The European Parliament has officially passed Extended Producer Responsibility (EPR) legislation that fundamentally shifts the responsibility for textile waste management to fashion brands and retailers – with far-reaching global implications. This new law requires all producers, including e-commerce platforms, to cover the full cost of collecting, sorting, and recycling textiles, regardless of whether they are based within or outside the EU. The financial burden of Europe's textile waste now falls squarely on the brands that create it. What are the critical business implications? UNIVERSAL SCOPE: The legislation applies to all producers selling in the EU market, including those of clothing, accessories, footwear, home textiles, and curtains. No company is exempt based on location. FAST FASHION PENALTY: Member states must specifically address ultra-fast and fast fashion practices when determining EPR financial contributions, creating cost penalties for unsustainable business models. GLOBAL SUPPLY CHAIN DISRUPTION: As the world's largest textile importer, the EU's new rules will ripple across global supply chains, particularly impacting exporters from Bangladesh, Vietnam, China, and India who supply much of Europe's fast fashion. TIMELINE PRESSURE: Officially adopted September 2025, this creates immediate operational and financial planning requirements. COMPETITIVE RESHAPING: Brands and retailers will inevitably pass increased costs down their supply chains, fundamentally altering supplier relationships and pricing structures globally. What are the implications for various stakeholders? For CEOs and board members: This represents more than regulatory compliance – it's a complete business model transformation. Companies must now integrate end-of-life costs into product pricing, rethink supplier partnerships, and accelerate circular design strategies. For sustainability and decarbonisation executives: This creates unprecedented opportunities for circular economy solutions, sustainable material innovation, and traceability system development across global supply chains. Link: https://lnkd.in/dTyHtHuD #sustainablefashion #circulareconomy #textilwaste #epr #fashionindustry #sustainability #supplychainmanagement #fastfashion #environmentalregulation #businessstrategy #decarbonisation #textilerecycling #fashionceos #boardgovernance #climateaction #wastemanagement #producerresponsibility #fashionsustainability #textileindustry #greenbusiness

Every time a card payment is processed, 𝘁𝗵𝗿𝗲𝗲 main types of fees are involved. Here’s a simple breakdown of the Three Core Fees: 1️⃣ Interchange Fee This is paid by your acquiring bank (or payment processor) to the cardholder’s bank (the issuer). It’s set by the card networks (like Visa and Mastercard; sometimes regulated), and is designed to cover things like fraud, credit losses, and infrastructure costs. 2️⃣ Scheme Fee Charged by the card networks themselves, this fee covers the operation of the payment system (“rails” that process the transaction). 3️⃣ Acquirer Markup This is the fee your acquirer or payment service provider (PSP) charges you, the merchant. It includes their costs, risk management, and profit margin for processing and settling the payment. The total cost a merchant pays is called the Merchant Service Charge, which is the sum of these three components. The Main Pricing Models: ► Bundled Pricing All fees are grouped into one flat rate. This is very common with small businesses. It’s easy to understand but doesn’t provide insight into what you’re actually paying for. ► Interchange+ The interchange fee and the acquirer’s fee are shown separately, but the scheme fee is typically bundled with the markup. This model offers some transparency. ► Interchange++ Each fee—the interchange, scheme, and acquirer markup—is itemized separately. This is the most transparent model and is favored by larger or multi-country merchants who want to track costs precisely. Who Chooses the Pricing Model? Most acquirers and PSPs decide what pricing model you’re offered. Unless you negotiate or have significant transaction volume, you’re likely to get bundled pricing by default. Larger or more experienced merchants who understand payments often push for Interchange++ for its clarity and fairness. Smaller merchants often aren’t aware that alternatives exist or find it difficult to compare offers. How Interchange Fees Vary Globally: Some regions (like the EU, UK, China, and Brazil) cap interchange fees to lower costs for merchants and stimulate competition. The US regulates only part of the system—such as capping debit card fees for large banks (the Durbin Amendment)—while credit card interchange remains uncapped and usually higher. Other countries, like India and Brazil, regulate interchange as part of broader financial inclusion goals. In markets with stricter regulation, merchants often benefit from lower, more predictable fees, making it easier to accept cards. Where fees are higher and less regulated, issuers can offer consumers more rewards (like cashback), but those costs are passed back to merchants—and sometimes their customers. Every model shifts the balance of costs and benefits between banks, merchants, and consumers in different ways. More info below👇, and I highly recommend reading my complete deep dive article about Interchange Fee and what factors impact the rate: https://bit.ly/44T4VJA

I frequently see conversations where terms like LLMs, RAG, AI Agents, and Agentic AI are used interchangeably, even though they represent fundamentally different layers of capability. This visual guides explain how these four layers relate—not as competing technologies, but as an evolving intelligence architecture. Here’s a deeper look: 1. 𝗟𝗟𝗠 (𝗟𝗮𝗿𝗴𝗲 𝗟𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗠𝗼𝗱𝗲𝗹) This is the foundation. Models like GPT, Claude, and Gemini are trained on vast corpora of text to perform a wide array of tasks: – Text generation – Instruction following – Chain-of-thought reasoning – Few-shot/zero-shot learning – Embedding and token generation However, LLMs are inherently limited to the knowledge encoded during training and struggle with grounding, real-time updates, or long-term memory. 2. 𝗥𝗔𝗚 (𝗥𝗲𝘁𝗿𝗶𝗲𝘃𝗮𝗹-𝗔𝘂𝗴𝗺𝗲𝗻𝘁𝗲𝗱 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻) RAG bridges the gap between static model knowledge and dynamic external information. By integrating techniques such as: – Vector search – Embedding-based similarity scoring – Document chunking – Hybrid retrieval (dense + sparse) – Source attribution – Context injection …RAG enhances the quality and factuality of responses. It enables models to “recall” information they were never trained on, and grounds answers in external sources—critical for enterprise-grade applications. 3. 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 RAG is still a passive architecture—it retrieves and generates. AI Agents go a step further: they act. Agents perform tasks, execute code, call APIs, manage state, and iterate via feedback loops. They introduce key capabilities such as: – Planning and task decomposition – Execution pipelines – Long- and short-term memory integration – File access and API interaction – Use of frameworks like ReAct, LangChain Agents, AutoGen, and CrewAI This is where LLMs become active participants in workflows rather than just passive responders. 4. 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗔𝗜 This is the most advanced layer—where we go beyond a single autonomous agent to multi-agent systems with role-specific behavior, memory sharing, and inter-agent communication. Core concepts include: – Multi-agent collaboration and task delegation – Modular role assignment and hierarchy – Goal-directed planning and lifecycle management – Protocols like MCP (Anthropic’s Model Context Protocol) and A2A (Google’s Agent-to-Agent) – Long-term memory synchronization and feedback-based evolution Agentic AI is what enables truly autonomous, adaptive, and collaborative intelligence across distributed systems. Whether you’re building enterprise copilots, AI-powered ETL systems, or autonomous task orchestration tools, knowing what each layer offers—and where it falls short—will determine whether your AI system scales or breaks. If you found this helpful, share it with your team or network. If there’s something important you think I missed, feel free to comment or message me—I’d be happy to include it in the next iteration.

The inventor of the SAFE note Adeo Ressi just eliminated the $150,000 and 6-month tax on starting a VC fund. This is huge, so we need to talk about it. Traditionally: ⏱️ Time: Launching a fund can take 6-12 months from thesis to first investment. 💸 Money: The VC setup cost ranges from $50,000 to $150,000+, with annual operations adding another $50,000+. 😵💫 Complexity: Requires three separate entities (LP, GP, and ManCo), complex legal agreements, and multiple regulatory filings. 🏦 Fund Size: There is a minimum fund size averaging $10M to make the fund economically viable. Each LP typically needs to invest $100K+ minimum because smaller checks are unprofitable due to per-LP administrative costs. 📊 Track Record: In order to raise this type of fund, new managers need larger LPs, and these larger LPs often need to see an existing successful investment track record, which some new managers don't have. These barriers have created a venture ecosystem where only those with established networks, significant resources, and/or institutional backing can participate. In 2025: Adeo came up with the Start Fund, a vehicle addressing all of the above head-on: ⏱️ Time: Set up a fund in ONE DAY vs. 6-12 months. 💸 Money: ZERO setup fees vs. $50K-$150K+. 😵💫 Complexity: ONE Delaware series vehicle vs. three separate entities, with an LPA just 1/3 the size. 🏦 Fund Size: Viable with just $250K+ vs. $10M minimum, and can accept smaller LPs (as low as $25K) because administration is streamlined 📊 Track Record: Fully portable track record that counts as fund one when you move to fund two. The benefits for emerging managers are clear: the barriers to entry are lower, giving a much wider pool of candidates a chance to create impact and shape the future. But here's why this matters for... LPs - The Start Fund allows LPs to participate with smaller check sizes, making it easier to diversify their portfolio - More of their capital actually goes to startups rather than overhead fees Startups: - This means more availability of capital from a wider range of sources - Access to a more diverse pool of venture investors with specialized expertise The Start Fund could fundamentally could change WHO gets to allocate capital to the next generation of startups, and WHO will benefit financially from it. I want to know what you all think. ------------- ✍️ Myrto Lalacos Follow for more content on launching and investing in VC firms

Open Banking (OB) isn’t a feature - it’s the blueprint for banks to stay relevant in an APIsed economy. But exposing a few APIs is not innovation. Here's what really powers OB - and some myth busting. OB is reshaping how we access and interact with financial services. At its core, it’s about unlocking data and making it securely available through modern infrastructure rails called APIs. But the impact goes far beyond banking. OB is becoming the key enabler of today’s two most dominant business models: —   Platform economics —   Embedded finance Banks play a critical role in this shift - because they hold the data. Enter: 𝗢𝗽𝗲𝗻 𝗕𝗮𝗻𝗸𝗶𝗻𝗴 𝗔𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲 This is the invisible technical foundation that allows banks to expose data and services to fintechs and partners. Here’s a simplified breakdown of key components: 1. API Gateway – The secure front door that handles requests and and routes them properly. 2. Consent & Identity Management – Ensures only the right parties get access, with the customer’s permission. 3. Authentication Layer – Uses secure login methods to confirm the customer’s identity. 4. Developer Portal – A gateway where third parties discover, test, and onboard to the bank’s APIs. 5. Microservices Layer – Breaks banking functions into modular services for faster, flexible delivery. 6. Core System Integration – Connects modern APIs to banks’ legacy systems without needing to rebuild everything from scratch. This isn’t just about technology - it’s about designing trust at scale. 𝗛𝗼𝘄 𝗮𝗻 𝗢𝗽𝗲𝗻 𝗕𝗮𝗻𝗸𝗶𝗻𝗴 𝗿𝗲𝗾𝘂𝗲𝘀𝘁 𝘄𝗼𝗿𝗸𝘀: 1.     A licensed third-party provider (TPP) sends an API request to the bank to access account data or initiate a payment. 2.     The end-user is redirected to the bank’s interface to authenticate and provide consent. 3.     Once consent is verified, the bank issues a secure access token to the TPP. 4.     The TPP retrieves only the authorized data or completes the payment transaction. 5.    All actions are logged for traceability, audit, security and compliance purposes. 𝗪𝗵𝗮𝘁’𝘀 𝗵𝗼𝗹𝗱𝗶𝗻𝗴 𝗯𝗮𝗻𝗸𝘀 𝗯𝗮𝗰𝗸? 1. Legacy tech – Many core platforms were never built for external connectivity. 2. Security & compliance pressure – Exposing APIs while meeting regulatory requirements is complex. 3. Real-time readiness – Open Banking requires real-time availability and minimal downtime. 4. Governance and ecosystem management – Managing third-party access and maintaining oversight is operationally demanding. Banks should avoid treating OB as just a tech upgrade or a compliance checkbox. It’s a strategic opportunity to modernize infrastructure - something they would have to do anyway. In the era of AI and real-time digital ecosystems, not being able to communicate via APIs is like owning a smartphone without internet access. Opinions: my own, Graphic source: Blanc Labs 𝐒𝐮𝐛𝐬𝐜𝐫𝐢𝐛𝐞 𝐭𝐨 𝐦𝐲 𝐧𝐞𝐰𝐬𝐥𝐞𝐭𝐭𝐞𝐫: https://lnkd.in/dkqhnxdg

If long-end bond yields spiral out of control, the Fed could start injecting liquidity again: a step-by-step guide of how it works. When a few weeks ago 30-year bond yields briefly flirted with the 5% level, the Fed's Collins released an interview stating that ''the Fed is absolutely ready to stabilize markets''. To stabilize the bond market, they would ''inject liquidity'' through operations like the LSAP - Large Scale Asset Purchase or QE. Central Banks create bank reserves when they perform such operations. Bank reserves are often referred to as ''Liquidity''. When Central Banks engage in liquidity creation, they do that in the hope that it activates the so-called Portfolio Rebalancing Effect. To understand this, let’s start from what QE does to the balance sheet of a commercial bank - take a look at the chart below. Following the GFC, regulators forced banks to own more HQLA (high quality liquid assets) to meet depositor outflows. Bank reserves and bonds qualify as ''HQLA'' as they are liquid enough to be converted in cash to meet potential outflows quickly. But banks are not indifferent between owning bank reserves and bonds, and especially if the amount of reserves grows dramatically as a result of QE. Bank reserves are a zero-duration and low-yielding instrument which can be suboptimal to own in big sizes especially if compared with bonds which offer higher returns and duration hedging properties. And this is when the Portfolio Rebalancing Effect kicks in. Once QE starts, Central Banks take away bonds and inject new reserves in the banking system. Loaded with suboptimal reserves, banks will try to switch back the composition of their portfolios towards more bonds. They will bid up safer bonds first, and bid up riskier bonds later when the hunt for returns intensifies. This will kick in a virtuous cycle of low volatility and a hunt for riskier assets: the Portfolio Rebalancing Effect in action. Summarizing: 1️⃣Central Banks expand their balance sheet and purchase bonds 2️⃣Commercial Banks are on the receiving end of QE, and hence their portfolio composition tilts towards more reserves, and less bonds; 3️⃣But reserves are sub-optimal to own compared to regulatory-friendly bonds, and hence they look to rebalance their portfolios; 4️⃣They start buying the very same bonds QE is buying, hence suppressing volatility further and compressing credit spreads; 5️⃣Asset allocators and investors across the world are more and more encouraged to take additional risks in their portfolio, supporting the flow of credit and capital. Does the Portfolio Rebalancing Effect make sense to you? 👉 If you enjoyed this post, follow me (Alfonso Peccatiello) to make sure you don't miss my daily dose of macro analysis.