PAX Gold (PAXG): Digital Gold's Safe Haven in the 2025 Crypto Storm

Page 10

‍

• Custody/Storage Fees: Paxos charges no ongoing custody fee for holding the gold that backs PAXG (Paxos | Pax Gold (PAXG)). This is a big selling point. Traditional vault storage and gold ETFs have annual fees (GLD charges ~0.40% per year in expense ratio to cover storage and management (Paxos | Pax Gold (PAXG))). Paxos currently forgoes such fees, likely for competitive reasons, and maybe because they cover costs through the transaction fees and other business lines. They possibly initially intended the transfer fee (0.02%) to function somewhat like a small annual fee as gold moves around, but once that was scrapped, PAXG effectively became cost-free to hold. This means long-term holders of PAXG don’t see their balances diminish at all – if you hold 100 PAXG for 5 years, you still have 100 PAXG (meanwhile if you held shares of GLD, the share count doesn’t drop but the NAV per share slowly drifts down vs spot because of fees). Paxos likely funds storage out of the cut they get from initial sales and possibly interest from clients’ fiat (if any held short-term). Gold storage is not hugely expensive (maybe 0.1-0.2% per annum of value in institutional vault costs). If PAXG scale increases, Paxos might revisit storage fees, but given "no custody fee" is part of their marketing comparison (Paxos | Pax Gold (PAXG)) (Paxos | Pax Gold (PAXG)), they seem committed to it for now.
  
  
• On-chain Transfer Fees: Initially, Paxos included a fee on on-chain transfers (0.02%) that would route to a fee beneficiary address (Paxos) (ARC: Add PAX Gold (PAXG) Collateral & Borrow Support - New Asset - Aave). This would have effectively been a small toll on any movement of PAXG between addresses, capturing value from secondary market activity too. However, due to incompatibility with DeFi and user pushback, Paxos removed this by setting it to zero (Paxos | Pax Gold (PAXG)) (Paxos | Pax Gold (PAXG)). Now, moving PAXG on-chain is free from Paxos’s perspective; users only pay Ethereum gas. Thus, Paxos no longer earns revenue from token transfers – only from the creation/redemption events that touch their system. This was a strategic trade-off: they gave up a potential continuous revenue (0.02% per transfer isn’t much unless velocity is extremely high, which for gold likely wasn’t huge anyway) in exchange for broader adoption and integration (which ultimately results in more creation/redemption as adoption grows).
  
  
• Other Potential Fees: If a user wants physical delivery, Paxos likely passes through the shipping/handling costs (armored transport, etc.) to the user. That’s not exactly revenue, more reimbursement. But they might charge a slight commission for arranging it. Alpha Bullion redemptions presumably come with markups on small bars/coins that Alpha (or Paxos via Alpha) profits from. Those details aren’t explicitly published, but one can assume it’s in line with typical bullion dealer premiums (a few percent on small items). Paxos proper might not take a cut of that beyond their standard fees; it may be mostly Alpha’s domain. Regardless, such cases are niche and not a primary revenue stream.
  
  

So Paxos’s revenue model for PAXG is primarily transaction-based. This is similar to an exchange or broker model rather than a continuous management fee model. The upside is that it doesn’t deter holding (no annual drag), encouraging users to keep PAXG long-term (which is good for adoption). The downside for Paxos is that if PAXG holders become very passive (buy and hold without trading or new issuance), Paxos’s direct revenue from PAXG would taper until those holders eventually sell.

But Paxos likely views PAXG as part of a suite that attracts clients to their platform, and there are indirect revenues: PAXG holders might also use Paxos’s exchange (itBit) for trading where Paxos charges trading fees. Or they might use Paxos’s custody or stablecoin services. Also, more PAXG in circulation could lead to more secondary trading which drives arbitrage creation/redemptions occasionally (each time generating some fee). Also, high supply builds brand value which could lead to partnerships that generate revenue.

Comparison to Competitors:

• Tether Gold (XAUt) charges a flat 25 bps (0.25%) each on buy and sell (and requires a minimum of 50 XAUt for redemption) (Tether Gold - The World's Leading Gold Token) (Tether Gold | XAUt token | Digital Token Backed by Physical Gold). Paxos’s fees range from 100 bps down to 12.5 bps depending on size, so for large transactions Paxos is much cheaper than Tether, but for very small ones Tether (25 bps) might be cheaper than Paxos’s 100 bps. However, small clients usually just trade on exchanges where fees are minimal anyway. Paxos’s tiered system is more inclusive for whales.
  
  
• Gold ETFs: GLD 40 bps/yr, others 15-30 bps/yr. Over many years, PAXG might be cheaper since Paxos doesn’t take annual fees, just one-time. For instance, holding GLD 5 years at 0.4% costs ~2% of value. Buying PAXG direct from Paxos might cost 0.25% one-time (if doing a large amount) and 0% holding = far cheaper in long run. This is a selling point Paxos highlights (Paxos | Pax Gold (PAXG)) (Paxos | Pax Gold (PAXG)). For frequent traders, an ETF might be cheaper because creation/redemption of ETF is done by market makers and end-user just pays small bid/ask spreads. But PAXG’s bid/ask is also tight on exchanges, so trading cost is low too.
  
  

Implications for Investors:

• When acquiring PAXG in size, it’s important to approach through the right channel to minimize fees (e.g., large investors should go direct to Paxos to get the 0.125% rate, whereas buying that much on an exchange might move the price and implicitly cost more than the fee).
  
  
• For small retail, it's often best to buy on an exchange where fees are just trading fees (like 0.1% on Binance) rather than 1% via Paxos direct. Paxos acknowledges this: they encourage trading on their itBit exchange or elsewhere for those who want to avoid creation fees (PAX Gold Fees – Paxos). They only charge their fees when converting through their wallet service.
  
  
• Redeeming small amounts directly might not make sense due to the 0.02 PAXG minimum fee (if you only have 1 PAXG, 0.02 is a 2% fee!). Those users should sell on exchange instead. Paxos’s model thus pushes very small holders to secondary markets and reserves direct dealing for larger customers – which is logical given overhead.
  
  

Sustainability of Revenue Model: Paxos’s fee structure seems sufficient to cover:

• Custody costs (perhaps ~0.1%/yr, but they get ~0.125-1% upfront, which can cover multiple years).
  
  
• Insurance and audits (again a fraction of a percent of assets probably).
  
  
• Overhead and profit. Since gold turnover might be on the order of 10-20% of supply per year (just a guess – meaning each year a fraction of holders churn), Paxos would capture new fees from that fraction. Combined with new growth, revenue is presumably healthy. Paxos also likely lends out (under strict conditions) any fiat held temporarily (though for PAXG, fiat is transitory, as they quickly convert to gold). They could also earn a tiny yield if they manage unallocated gold for short periods, but likely not – they match tokens to allocated quickly.
  
  

Other revenue aspects: Indirectly, PAXG helps Paxos’s overall business (maybe those customers also use USDP or their crypto brokerage). But focusing on PAXG: Paxos can be profitable with a relatively modest user base because each PAXG creation has a few basis points margin plus there's no continuous servicing cost beyond vault fees.

If PAXG volume increased dramatically, Paxos might consider reducing fees to stay competitive (maybe flattening the tier further). For example, if a competitor came with 0.1% fees for all sizes, Paxos might adjust. As of now, the fee structure hasn’t changed publicly, except effectively lowered via removal of transfer fee.

In summary, PAXG’s fee and revenue model is simple and consumer-friendly (no hidden costs) and gives Paxos a sustainable income stream aligned with usage. It eliminates holding costs for users, thereby potentially attracting long-term holders. For analysis, one might estimate Paxos’s revenue from PAXG: e.g., in 2024 if 50k PAXG were created anew and 30k redeemed, at an average fee 0.2-0.3%, Paxos might earn on order of 80k * 0.25% = 200 PAXG ($600k if gold $3000). Not enormous, but as adoption grows and if millions of PAXG eventually circulate with churn, this can scale. Considering Paxos’s entire business (stablecoins etc.), PAXG is one stream but not the only one. They might accept the lower fees on PAXG as part of a holistic strategy to be the go-to tokenization platform.

For investors, the key takeaway is that friction costs around PAXG are low and transparent, and Paxos’s interests are aligned with token holders (they want more PAXG circulation and value, since they earn when users enter/exit). There’s no scenario where Paxos would need to impose additional fees unexpectedly; if anything, fees might go down over time if economies of scale allow, which would benefit holders further.

Holding Costs and Yield Potential

By design, holding PAXG is meant to mimic holding physical gold – meaning it doesn’t inherently produce yield or cash flows. However, the crypto ecosystem around PAXG can enable yield generation that physical gold normally wouldn’t provide. Let’s break this into two aspects: cost of carry (holding costs) and yield opportunities.

Holding Costs / Cost of Carry: One of the attractive features of PAXG is the lack of direct holding fees (as discussed above). For an investor who simply holds PAXG in their wallet, there is no custody fee, no decay of tokens. Compare this to:

• Holding physical gold: you might pay storage fees or insurance annually (if storing in a vault or safe deposit box). If holding at home, you avoid fees but have risk/security costs.
  
  
• Holding a gold ETF: you pay the fund’s expense ratio (GLD ~0.4%/yr, IAU 0.25%/yr, etc.), which is indirectly reflected in the ETF’s price performance versus spot. PAXG thus has effectively a zero net holding cost from Paxos’s side. One could say Ethereum wallet maintenance (gas for any token moves) is a minor overhead, but if just holding, you incur none.
  
  

In classical terms, the cost of carry for gold is typically considered the storage cost minus any convenience yield. PAXG reduces storage cost to near zero and arguably provides a convenience yield (ease of use) rather than subtracting one. This could imply that PAXG might sometimes trade at a slight premium to gold because it eliminates these carrying costs. Indeed, if GLD holders pay 0.4% per year, one could arbitrage by holding PAXG and shorting GLD, expecting GLD to underperform. However, transaction frictions and regulatory differences likely limit that arbitrage.

Because Paxos doesn't charge holding fees, we should consider if inflation of supply could be a factor (like Tether prints new USDT to invest reserves – not applicable here since Paxos can’t “invest” gold; the gold just sits). Gold itself has an inflation (mining adds ~1-2% supply per year globally, which can put mild downward pressure on price if not matched by demand). But holding PAXG, you indirectly face gold’s supply inflation like any gold investor – no different.

Taxes are an external holding cost: in some jurisdictions, holding gold (and thus PAXG) has wealth tax or when sold is taxed as a collectible (like in US, gold gains are taxed at 28% cap gain). That’s not unique to PAXG but an investor consideration.

Yield Potential: While PAXG itself doesn’t pay interest, the crypto environment allows holders to generate yield by lending or staking PAXG in various ways:

• Centralized Lending: Platforms such as Nexo, Celsius (before its demise), YouHodler, etc., have offered yields on PAXG deposits – often in the range of 3-8% APY (Get An Instant Paxos Gold Loan - Borrow PAXG Instantly Online) (Earn Free Paxos Gold - Earn yield on Paxos Gold Up to 7%). How can they pay that if gold yields 0? It’s because they lend out your PAXG to borrowers (traders or institutions) who pay interest to borrow gold. Who would borrow gold? Possibly traders who want to short gold in the crypto markets or arbitrageurs betting on price differences. It could also be market makers needing inventory on exchanges to facilitate trades. The rates might be temporarily high if, for example, someone is willing to pay to short gold during a rally. However, these yields are not risk-free – you rely on the platform’s solvency and risk management (Celsius is a cautionary tale where depositors lost assets).
  
  
• DeFi Lending Pools: Protocols like Compound and Aave either already support or are evaluating supporting PAXG (Pax Gold (PAXG): Key Information - Coinhouse). If PAXG is listed on a major DeFi lending market, holders can supply PAXG and earn interest from borrowers. As of now, Compound has PAXG (with typically low utilization – meaning low interest because few borrow). If Aave onboards PAXG, interest might pick up if there’s demand to borrow PAXG (which might come from traders or possibly gold arbitrage strategies).
  
  
• Yield Farming and Liquidity Providing: PAXG can be paired in liquidity pools on DEXes (e.g., PAXG/ETH or PAXG/USDC on Uniswap). Liquidity providers earn trading fees, and sometimes incentives. For example, if a DEX or aggregator offered rewards for a PAXG pool, LPs could get some yield. Even without extra incentives, the 0.3% swap fee on Uniswap for volatile pairs can translate to some APY if volume is decent. For a PAXG/USDC pool (more stable-stable like behavior if gold stable relative to USD short-term), fees might be lower but still some yield relative to exposure. However, LPs take on impermanent loss risk if gold price moves significantly relative to the paired asset. For instance, in a PAXG/USDC pool, if gold price surges, LP could end up with more USDC and less PAXG, missing some upside (and vice versa if gold falls). So risk-averse holders might not LP because they want pure gold exposure.
  
  
• Staking in DeFi Protocols: Some strategies might involve using PAXG as collateral on MakerDAO to mint DAI, then investing that DAI somewhere, effectively leveraging gold. While not a direct yield on PAXG, one can amplify returns by borrowing against PAXG. For example, if one thinks gold is stable or rising, they might borrow stablecoins at low cost (Maker’s stability fee maybe a few percent) and then use those to earn yield or buy more gold. This is like a carry trade using gold as collateral. It's complex and adds risk (liquidation if gold drops too much), but it's an option.
  
  
• Interest through Derivatives: While not exactly yield, an advanced user could lend PAXG on margin to short-sellers on an exchange like Kraken (if margin trading for PAXG is enabled), earning interest from margin fees. Or they could write covered calls on PAXG (if an options market existed) to earn premium (though no big options on PAXG yet; one could simulate with gold futures and PAXG).
  
  

It’s worth noting that traditional gold has almost zero yield opportunities outside of maybe leasing gold in the OTC market (which is typically low yield ~0.5% annual for gold lease rates in normal times, and mainly accessible to bullion banks). PAXG democratizes gold lending – now any holder can potentially lend and earn interest. This is a significant innovation: effectively turning a non-productive asset into one that can generate cash flow if you’re willing to take some risk.

Risks and Considerations for Yield: Earning yield on PAXG means taking on additional risks:

• Counterparty risk (if you lend via centralized platform).
  
  
• Smart contract risk (if via DeFi).
  
  
• Liquidity/volatility risk (if LPing or using as collateral). So an investor must weigh the ~3-5% possible yields against these risks. But at least the opportunity is there, whereas with physical gold under your mattress, you get 0% and maybe pay for storage or insurance.
  
  

Paxos and Yield: Paxos itself does not offer a yield on PAXG accounts (they aren’t lending out customer gold – that would violate the full-reserve principle and trust status). So any yield is externally driven. Paxos did partner with Genesis Lending in 2020 to let users lend PAXG (Genesis would borrow it) (Put Your Gold to Use! PAX Gold Now Available through Kraken ...). That program likely offered some interest. However, with Genesis’s troubles in 2022 (they had issues due to broader crypto credit crunch), that might have paused. Nonetheless, this shows Paxos has explored enabling yield in a safe way (Genesis was a big, institutional borrower, though even they faced issues eventually).

Opportunity Cost: One way to evaluate holding PAXG is opportunity cost. If one holds USD, they can earn interest (today, USD stablecoins can yield a few percent in DeFi or Treasuries yield ~5%). If one holds PAXG, they only get gold's price return (~24% YTD 2025 rally but could be flat or down some years) and if they want interest they have to lend it. So, if real interest rates are high, the opportunity cost of holding gold/PAXG (which yields 0 by itself) rises. This is classic for gold: when interest rates rise, gold often struggles because investors prefer yield assets (Gold ETF Inflows Hit Three-Year High as PAXG, XAUT Outperform Wider Crypto Market). But PAXG’s presence in DeFi partly counters this: you could hold PAXG and supply to Aave to earn, say, 2% APY if people borrow it, cutting the opportunity cost gap. Not as good as holding cash at 5%, but better than zero.

‍

Thank you for taking the time to read this article. We invite you to explore more content on our blog for additional insights and information.

https://www.thestandard.io/blog  

‍

"If you have any comments, questions, or suggestions, please do not hesitate to reach out to us at [ https://discord.gg/K72hed6FRE ]. We appreciate your feedback and look forward to hearing from you."

‍

CLICK HERE TO CONTINUE

PAGE 11: www.thestandard.io/blog/pax-gold-paxg-digital-golds-safe-haven-in-the-2025-crypto-storm-11