Has the AI Crash Already Begun? A Clear-Eyed Look at the Numbers

Over a trillion dollars evaporated from tech stocks in a single week this summer. Then it happened again in early July. Chipmakers in Seoul got hit so hard that the Korea Exchange had to halt trading twice just to stop the bleeding. Somewhere between those two moments, a lot of very smart people started asking the same question out loud: Is this the start of the crash, or just a very loud correction inside a boom that still has legs?

We are not going to give you a comforting answer here. We are going to give you the actual numbers, the actual warnings from people who run trillion-dollar institutions, and a framework for figuring out which camp you should be in. Because right now, both camps have real evidence. That is exactly what makes this moment dangerous to get wrong.

The Week the Market Panicked

In late June 2026, the Kospi index closed down nearly 8 per cent in a single session. SK Hynix lost close to 15 per cent. Samsung Electronics fell more than 9 per cent. Combined, the two companies erased close to $290 billion in value in one trading day. Consequently, the Korea Exchange suspended the program to slow the freefall.

The pain did not stay in Asia. The Nasdaq slid 2.2 per cent that same week, and Micron Technology dropped 13 per cent after rising nearly 800 per cent over the prior year. By June 26, the Nasdaq had fallen 5 per cent for the month. Oracle, heavily leveraged into AI infrastructure, posted its worst week since the dot-com bubble burst. Meanwhile, Apple raised product prices, citing higher chip costs, which only fed the anxiety.

Then, right as July opened, it happened again. A weak June jobs report rattled markets, and memory and AI-related stocks tumbled further. Samsung and SK Hynix, which together now make up roughly half of the Kospi’s total weight, dragged the whole index down again. Hong Kong-listed chip firms fell in tandem, some by double digits. This is not a one-off. It is starting to look like a pattern.

What Actually Happened, In Plain Numbers

Let’s separate signal from noise. According to one market recap, the so-called Magnificent Seven fell roughly 12.7 per cent collectively in June. That is a real drawdown. However, the same analysis notes that Nvidia, Microsoft, and Alphabet kept reporting record data centre revenue throughout the selloff. Fundamentals and stock prices briefly stopped agreeing with each other. That gap is worth sitting with.

Here is the tension in one sentence: earnings kept climbing while stock prices fell. That combination usually means one of two things. Either the market is repricing risk ahead of an actual slowdown, or it is simply digesting years of runaway gains. Nobody can tell you with certainty which one it is yet. Anyone who claims otherwise is selling you something.

Event	Date	Magnitude
Kospi single-day drop	June 23, 2026	~8%, $290B lost across Samsung and SK Hynix
Nasdaq weekly decline	Week of June 23, 2026	2.2% single day, 5% by month-end
Micron Technology	June 2026	-13% after +800% trailing 12 months
Oracle’s weekly loss	Late June 2026	Worst week since the dot-com bust
Magnificent Seven, June	June 2026	-12.7% collectively

The Bear Case: Where the Cracks Are Real

Let’s start with the argument that this is genuinely the beginning of the end. It rests on a few pillars, and they are sturdier than the AI optimists want to admit.

The Productivity Gap Nobody Wants to Talk About

MIT’s Media Lab published a widely cited study in 2025 called The GenAI Divide. Despite $30 to $40 billion in enterprise generative AI spending, the researchers found that 95 per cent of corporate pilots delivered zero measurable return. Only 5 per cent of projects scaled into anything resembling meaningful revenue. That is a staggering failure rate for a technology that companies are betting entire balance sheets on.

To be fair, the methodology has critics. Marketing AI Institute founder Paul Roetzer argued the study undercounted efficiency gains that do not show up as direct revenue. That is a legitimate critique. Still, a separate National Bureau of Economic Research paper from February 2026 found that 90 per cent of firms reported no measurable productivity impact from AI, even as executives kept projecting rosy future gains. Two independent studies landing near the same number is hard to wave away.

The Circular Money Problem

Here is where things get genuinely strange. NVIDIA sells chips to cloud providers. Those providers build data centres, and rent compute back to AI labs. The AI labs burn that cash training models, then raise more money, often from the very same tech giants that sold them the chips in the first place. Money moves in a loop, and every step of that loop books it as revenue.

OpenAI illustrates the scale of this problem well. The company has committed to roughly $1.4 trillion in data centre spending over eight years, against annual revenue of around $13 billion. That spending is largely funded by debt, not existing cash flow. Morgan Stanley estimates global data centre spending will hit $3 trillion by 2028, with roughly half of that financed through private credit. When the underlying revenue does not match the scale of the commitment, somebody eventually has to eat the gap.

The Bull Case: Why This Isn’t 2000 All Over Again

Now let’s steelman the other side, because it deserves a fair hearing. Fidelity’s research desk points out that, unlike the dot-com years, today’s capex is largely funded from earnings rather than debt for most hyperscalers. Their capex-to-free-cash-flow ratio currently sits below 1, compared to nearly 4 times free cash flow at the dot-com peak. That is a meaningful structural difference.

Adoption also looks nothing like 1999. By 2025, roughly 71 per cent of organisations will regularly use generative AI in at least one business function, and OpenAI alone reports 800 million weekly active users. Compare that to the dot-com era, when most companies going public had no working revenue model at all, and the internet itself had a fraction of today’s user base. AI, whatever its flaws, is already embedded in daily workflows for hundreds of millions of people.

There is also the profitability point. Guinness Global Investors notes that, excluding Oracle, the other major hyperscalers can fund their entire forecast capex out of operating cash flow. Today’s Magnificent Seven trade at a median forward multiple around 26 times earnings, well below the extremes seen at previous bubble peaks. So the argument goes: yes, spending is enormous, but the companies doing the spending are real, profitable, and not (mostly) drowning in debt to do it.

Comparing the Dot-Com Bubble and the AI Boom

Numbers make this comparison easier to hold in your head than prose does. Here is where the two eras genuinely differ, and where they rhyme uncomfortably.

Metric	Dot-com peak (2000)	AI boom (2026)
Capex-to-sales ratio	~32%	~34%, projected 37% by 2028
Capex funding source	Heavy debt and equity issuance	Mostly operating cash flow (ex-Oracle)
Median forward P/E, top 7 stocks	Far above 30x	~26.1x
Business model maturity	Many pre-revenue startups	Profitable incumbents plus unprofitable labs
Real-world adoption	Low, dial-up era	71% of firms use GenAI regularly

Notice the pattern. On corporate balance sheets, this cycle looks healthier than 2000. On the specific question of whether AI software revenue justifies AI infrastructure spending, though, the gap is arguably worse. According to ETF Trends, hyperscaler infrastructure spending is running at nearly twenty times the combined revenue of pure-play AI software companies like OpenAI and Anthropic. That imbalance has to close somehow. The only real debate is whether it closes gradually through revenue growth or abruptly through write-downs.

The Concentration Risk Hiding in Plain Sight

A handful of companies are now doing most of the heavy lifting for entire stock indexes. In South Korea, Samsung and SK Hynix alone make up around half the Kospi’s total weight, up from roughly a quarter just a year earlier. That means a bad morning for two companies can drag down nine hundred other listed businesses that had nothing to do with AI chips.

Yale’s Chief Executive Leadership Institute flagged this exact dynamic. At a recent summit, over 150 top CEOs, including venture capitalists and consulting-firm partners, warned that a small cluster of companies is securing most major AI deals. If the bold promises behind those deals fall short, the tight web of dependency among OpenAI, Nvidia, Microsoft, Google, and a few others could trigger a chain reaction. Yale’s researchers explicitly compared the risk to the contagion dynamics of the 2008 financial crisis. That is not a comparison serious institutions make lightly.

Follow the Debt: Private Credit and the IMF’s Warning

This is the part of the story that gets the least airtime, and it might matter the most. Global private credit assets under management have grown from roughly $158 billion in 2010 to nearly $2.5 trillion today. A meaningful chunk of that has flowed into leveraged buyouts of software and tech services companies, priced on the assumption that revenue keeps growing indefinitely.

At the IMF’s spring 2026 meetings, officials fielded direct questions about whether private credit poses systemic risk tied to insurer exposure and business development companies. The concern is straightforward. Insurers hold large amounts of private credit paper. If AI-driven revenue assumptions prove too optimistic, and borrowers start missing covenants, the losses do not stay contained to venture investors. They ripple into insurance balance sheets that ordinary households depend on for annuities and life policies.

Private credit valuations also lag reality by design. Funds typically report net asset values with a 60 to 990-day delay, using model-based marks rather than live market prices. During periods of stress, public credit spreads widen months before private markets catch up. That creates a dangerous illusion of stability right up until it doesn’t. If you are looking for where a genuine AI-driven financial shock might originate, this opaque corner of the credit market is a better place to look than the stock ticker on your phone.

The Bill Nobody Budgeted For: Electricity

Here is the cost of the AI boom that shows up on regular people’s monthly bills, not just balance sheets. Between 2018 and 2023, data centres’ share of total U.S. electricity use rose from 1.9 per cent to 4.4 per cent. Utilities requested more than $29 billion in rate increases in just the first half of 2025, double the prior year’s pace. Consequently, average U.S. electricity prices climbed to about 19 cents per kilowatt-hour by the end of 2025, roughly 27 per cent higher than in 2019.

The regional impact is stark. In Virginia’s “Data Centre Alley,” electricity prices in high-concentration areas jumped 267 per cent over five years, according to a Bloomberg analysis. One resident described a monthly bill that nearly tripled without warning. Nearly three-quarters of Virginia voters now blame data centres for rising costs. Meanwhile, a peer-reviewed study projected that Virginia’s electricity generation costs specifically could rise as much as 57 per cent by the end of the decade. Lawmakers noticed. In 2026 alone, legislators in more than 30 states introduced over 300 bills addressing data centre energy policy, ranging from construction moratoriums in New York and South Dakota to new rate classes that force data centres to pay their own interconnection costs. California’s Little Hoover Commission went further, urging lawmakers to make tech companies, not households, foot the bill for grid upgrades. This regulatory backlash is not a minor side story. It is a direct political risk to the economics of every hyperscaler’s build-out plan.

South Korea’s Wake-Up Call

It’s worth zooming in on South Korea specifically, since it has become the market’s most sensitive nerve ending for AI anxiety. Samsung and SK Hynix supply the memory chips that power AI training clusters worldwide. When U.S. semiconductor stocks wobble, Korean markets amplify the move because so much of their national index rides on just two companies. That is precisely why the Kospi has now halted trading multiple times in a matter of weeks.

Ironically, this selloff arrived just days after South Korea’s government announced major state investment initiatives to support its AI ambitions, including a combined 800 trillion won commitment from Samsung and SK Hynix for a national semiconductor ecosystem project. SK Hynix even planned to begin trading American depositary receipts on the Nasdaq around the same time. The timing captures the whole moment perfectly: governments doubling down on AI infrastructure investment at the exact moment markets started asking hard questions about whether the returns will ever show up.

What the Bulls Get Right

We would be doing you a disservice if we pretended the optimists have nothing. Wells Fargo’s Ohsung Kwon put it bluntly in a note to clients: “AI keeps bubbling,” and until growth actually slows or inflation meaningfully accelerates, there may be limited downside. That is not blind optimism. It is a bet that momentum plus real earnings growth can coexist with stretched valuations for longer than sceptics expect. Analysts at Ark Invest have made a similar point: even with capex rising to dot-com-era levels, price-to-earnings ratios for today’s tech leaders remain far below those seen in 2000.

Oracle’s debt load, often cited as the canary in the coal mine, has defenders, too. Wedbush analyst Dan Ives argues the company’s “remaining performance obligations,” meaning future contracts not yet booked as revenue, total $553 billion against a much smaller capex-to-RPO ratio than peers. In his view, Oracle is building against real, already-signed demand rather than speculative hope. If that reading holds, at least one of the market’s biggest debt-fueled worries is less alarming than the headlines suggest.

Five Signals Worth Actually Watching

Instead of asking “has the crash begun” as a yes-or-no question, track these five signals over the coming quarters. They will tell you more than any single week’s headline ever could.

• Capex-to-revenue ratio for pure-play AI labs. Watch whether OpenAI and Anthropic’s revenue growth starts closing the roughly twenty-to-one gap with infrastructure spending, or whether it widens further.
• Private credit valuation gaps. Any sudden write-down in business development company marks, especially among tech-focused lenders, would be an early warning that the opaque part of this market is cracking.
• State-level data centre legislation. If moratoriums in New York, South Dakota, or Oklahoma actually pass and take effect, that directly constrains hyperscaler build-out timelines.
• Enterprise AI ROI data beyond MIT’s single study. Watch for corroborating or contradicting research from other institutions over the next year.
• Concentration ratios in national indexes. If Samsung and SK Hynix’s share of the Kospi keeps climbing, expect more violent single-day swings, regardless of what’s happening with fundamentals.

What This Means for Your Portfolio

We are not your financial advisor, and this is not a stock recommendation. That said, here is the honest framework worth carrying forward. The AI boom is not a Ponzi scheme built on nothing, unlike some of the dot-com era’s more absurd business plans. Real revenue exists. Real adoption exists. At the same time, the spending gap between infrastructure build-out and actual software revenue is historically wide, and a meaningful chunk of that spending runs through opaque, debt-heavy channels that do not reprice quickly when sentiment turns. Diversification across sectors, not just within tech, remains the boring but reliable answer. If a correction is coming, concentrated bets on the most hyped names will hurt the most. If the boom continues, broad exposure still captures plenty of upside without betting the house on a handful of tickers.

Whatever you decide, do it with your eyes open. The people running these companies, including Sam Altman himself, have publicly warned that investors will overinvest and lose money during this phase. When the person raising the money tells you that directly, it is worth listening.

Spend some time for your future. 

To deepen your understanding of today’s evolving financial landscape, we recommend exploring the following articles:

College Majors That Leave You Drowning in Debt (And the Ones That Don’t) 

The Smart Financial Plan: Order of Operations That Actually Works 

6 Broke-to-Billionaire Stories and the Habits That Made Them 

Why the Coming AI Crash Will Make the Global Financial Crisis Look Easy 

Explore these articles to get a grasp on the new changes in the financial world.

Disclaimer

This article is provided for general informational and educational purposes only. It does not constitute financial, investment, legal, or tax advice, and it should not be relied upon as the basis for any investment decision. Market conditions change rapidly, and figures cited here reflect publicly available reporting as of the time of writing. Readers should consult a licensed financial advisor, accountant, or attorney before making any decisions based on the information presented. The author and publisher accept no liability for losses arising from reliance on this content.

References

[1] Wikipedia contributors, “AI bubble,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/AI_bubble
[2] “When it all comes crashing down: The aftermath of the AI boom,” The Bulletin of the Atomic Scientists, Dec. 2025. [Online]. Available: https://thebulletin.org/2025/12/when-it-all-comes-crashing-down-the-aftermath-of-the-ai-boom
[3] “This Is How the AI Bubble Bursts,” Yale Insights, Yale School of Management. [Online]. Available: https://insights.som.yale.edu/insights/this-is-how-the-ai-bubble-bursts
[4] “Trillion-dollar AI market wipeout happened because investors banked that ‘almost every tech company would come out a winner,'” Fortune, via Yahoo Finance. [Online]. Available: https://finance.yahoo.com/news/trillion-dollar-ai-market-wipeout-115521847.html
[5] “MIT report: 95% of corporate generative AI pilots fail to deliver returns,” Computing, 2025. [Online]. Available: https://www.computing.co.uk/news/2025/ai/mit-report-95pc-corporate-generative-ai-pilots-fail
[6] “Beyond ROI: Are We Using the Wrong Metric in Measuring AI Success?” UC Berkeley Executive Education, 2025. [Online]. Available: https://exec-ed.berkeley.edu/2025/09/beyond-roi-are-we-using-the-wrong-metric-in-measuring-ai-success/
[7] “MIT Says 95% Of Enterprise AI Fail,” Forbes, Aug. 2025. [Online]. Available: https://www.forbes.com/sites/jaimecatmull/2025/08/22/mit-says-95-of-enterprise-ai-failsheres-what-the-5-are-doing-right/
[8] “MIT Report Finds Most AI Business Investments Fail,” Virtualisation Review, Aug. 2025. [Online]. Available: https://virtualizationreview.com/articles/2025/08/19/mit-report-finds-most-ai-business-investments-fail-reveals-genai-divide.aspx
[9] “MIT Report Finds 95% of AI Pilots Fail to Deliver ROI,” Legal.io, 2025. [Online]. Available: https://www.legal.io/articles/5719519/MIT-Report-Finds-95-of-AI-Pilots-Fail-to-Deliver-ROI-Exposing-GenAI-Divide
[10] “MIT report: 95% of generative AI pilots at companies are failing,” Fortune, Aug. 2025. [Online]. Available: https://fortune.com/2025/08/18/mit-report-95-percent-generative-ai-pilots-at-companies-failing-cfo/
[11] “That Viral MIT Study Claiming 95% of AI Pilots Fail? Don’t Believe the Hype,” Marketing AI Institute, 2025. [Online]. Available: https://www.marketingaiinstitute.com/blog/mit-study-ai-pilots
[12] “MIT: Why 95% of Enterprise AI Investments Fail to Deliver,” AI Magazine, Sept. 2025. [Online]. Available: https://aimagazine.com/news/mit-why-95-of-enterprise-ai-investments-fail-to-deliver
[13] “MIT report finds 95% of enterprises see no return on generative AI,” Digital Commerce 360, 2025. [Online]. Available: https://www.digitalcommerce360.com/2025/08/25/mit-report-no-return-on-generative-ai/
[14] “MIT Finds 95% Of GenAI Pilots Fail Because Companies Avoid Friction,” Forbes, Aug. 2025. [Online]. Available: https://www.forbes.com/sites/jasonsnyder/2025/08/26/mit-finds-95-of-genai-pilots-fail-because-companies-avoid-friction/
[15] “AI Bubble vs. Dot-com Bubble: A Data-Driven Comparison,” IntuitionLabs, Mar. 2026. [Online]. Available: https://intuitionlabs.ai/articles/ai-bubble-vs-dot-com-comparison
[16] “Is AI a bubble? 5 signs to watch for,” Fidelity, Feb. 2026. [Online]. Available: https://www.fidelity.com/learning-center/trading-investing/ai-bubble
[17] J. Eric, “AI Bubble 2026: Is It Real? Capex, Fed Warnings & GPU Lifespans,” Medium, Jun. 2026. [Online]. Available: https://medium.com/@svnkrmkr/ai-bubble-2026-is-it-real-capex-fed-warnings-gpu-lifespans-b5db2178d350
[18] Aftab, “The $2 Trillion AI Bubble Is About to Pop,” Medium, Apr. 2026. [Online]. Available: https://medium.com/@aftab001x/the-2-trillion-ai-bubble-is-about-to-pop-heres-exactly-how-it-happens-0dba6f7a6a1f
[19] “Bull vs. Bear: Is the AI Revolution Nearing a Dot-Com Correction?” ETF Trends, Apr. 2026. [Online]. Available: https://www.etftrends.com/thematic-investing-content-hub/bull-vs-bear-ai-revolution-nearing-dot-com-correction/
[20] “Are we in an AI bubble?” Guinness Global Investors. [Online]. Available: https://www.guinnessgi.com/insights/are-we-in-an-ai-bubble
[21] “AI Capex Boom Vs. The Dot-Com Bubble,” Seeking Alpha, Feb. 2026. [Online]. Available: https://seekingalpha.com/article/4868823-ai-capex-boom-vs-the-dotcom-bubble-the-striking-similarities-and-three-big-concerns
[22] “AI capex is a ‘euphoric’ bubble, and you should buy into it, Wells Fargo analyst urges,” Fortune, May 2026. [Online]. Available: https://fortune.com/2026/05/13/ai-euphoric-bubble-buy-it-wells-fargo/
[23] “July 2026 Stock Market Rally: Navigating AI-Driven Volatility,” Intellectia.AI. [Online]. Available: https://intellectia.ai/blog/july-stock-market-rally-ai-volatility-2026
[24] “Is AI ‘one big bubble’? Behind the tech sell-off,” NPR, Jun. 2026. [Online]. Available: https://www.npr.org/2026/06/23/nx-s1-5867633/ai-selloff-tech-stocks-bubble-nasdaq
[25] “AI, chip stocks slump as sell-off continues after global peers fall,” Seeking Alpha, Jul. 2026. [Online]. Available: https://seekingalpha.com/news/4609566-ai-memory-chip-stocks-dip-after-sk-hynix-samsung-drag-down-kospi
[26] “Samsung Electronics, SK Hynix shares tumble over 9% as chip rout spreads from Wall Street,” CNBC, Jul. 2026. [Online]. Available: https://www.cnbc.com/2026/07/02/samsung-sk-hynix-shares-slide-kospi-tech-selloff-nasdaq.html
[27] “South Korean Stocks Drop 6% as AI Concerns Weigh on Chipmakers,” Bloomberg, Jul. 2026. [Online]. Available: https://www.bloomberg.com/news/articles/2026-07-02/south-korean-stocks-tumble-6-as-ai-jitters-hurt-chipmakers
[28] “Data Centre Power Demands Are Contributing to Higher Energy Bills,” Environmental and Energy Study Institute. [Online]. Available: https://www.eesi.org/articles/view/data-center-power-demands-are-contributing-to-higher-energy-bills
[29] “AI Data Centres: Big Tech’s Impact on Electric Bills, Water, and More,” Consumer Reports, Mar. 2026. [Online]. Available: https://www.consumerreports.org/data-centers/ai-data-centers-impact-on-electric-bills-water-and-more-a1040338678/
[30] “U.S. Data Centre Power Consumption Map by State (2026),” Electric Choice. [Online]. Available: https://www.electricchoice.com/datacenters/
[31] “State Data Centre Legislation in 2026 Tackles Energy and Tax Issues,” MultiState, Feb. 2026. [Online]. Available: https://www.multistate.us/insider/2026/2/20/state-data-center-legislation-in-2026-tackles-energy-and-tax-issues
[32] “AI data centres could hike California electricity bills,” CalMatters, Mar. 2026. [Online]. Available: https://calmatters.org/environment/2026/03/little-hoover-data-center-electricity/
[33] “2026 AI Data Centre Bill Watch List,” Citizens Action Coalition, Feb. 2026. [Online]. Available: https://www.citact.org/news/2026-ai-data-center-bill-watch-list
[34] “With electricity bills rising, some states consider new data centre laws,” Stateline, Feb. 2026. [Online]. Available: https://stateline.org/2026/02/05/with-electricity-bills-rising-some-states-consider-new-data-center-laws/
[35] “Electricity prices are up 40% since 2021, but data centres shouldn’t get all the blame,” Fortune, May 2026. [Online]. Available: https://fortune.com/2026/05/20/electricity-bills-surging-not-just-data-centers/
[36] “Americans’ AI hate wave might just be gathering steam,” Fortune, May 2026. [Online]. Available: https://fortune.com/2026/05/19/data-centers-electricity-costs-us-public-opinion/
[37] International Monetary Fund, “Press Briefing Transcript: Global Financial Stability Report, Spring Meetings 2026,” Apr. 2026. [Online]. Available: https://www.imf.org/en/news/articles/2026/04/15/tr-04142026-press-briefing-transcript-global-financial-stability-report-spring-meetings-2026
[38] “IMF Warns AI Has Made Cyber Risk a Financial Stability Threat,” BankInfoSecurity, May 2026. [Online]. Available: https://www.bankinfosecurity.com/imf-warns-ai-has-made-cyber-risk-financial-stability-threat-a-31679
[39] Wikipedia contributors, “Dot-com bubble,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/Dot-com_bubble
[40] Wikipedia contributors, “Nvidia,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/Nvidia
[41] Wikipedia contributors, “OpenAI,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/OpenAI
[42] Wikipedia contributors, “Nasdaq Composite,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/Nasdaq_Composite
[43] Wikipedia contributors, “S&P 500,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/S%26P_500
[44] Wikipedia contributors, “Federal Reserve,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/Federal_Reserve
[45] Wikipedia contributors, “Financial Stability Board,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/Financial_Stability_Board
[46] Wikipedia contributors, “Private credit,” Wikipedia, The Free Encyclopedia. [Online]. Available: https://en.wikipedia.org/wiki/Private_credit
[47] International Monetary Fund, official website. [Online]. Available: https://www.imf.org