Forex Swaps

A Question You Might Ask About PayPal

A question arose in a macroeconomics forum that runs in my head. From an MMT lens a person in say New Zealand never needs to “get the SGD or resp. the USD” to buy goods from Singapore resp. USA. A good proof might be PayPal. All that happens is that behind the scenes a forex swap transaction is made, and the PayPal user is charged a few (does not get 100% of the donors $ amount wired via PayPal.)

Q.1 How does PayPal handle these swaps?

Both,
(a) When I purchase foreign goods using PayPal?
(b) When I transfer a paypal amount to my domestic bank account?

Q.2 Obviously state departments do not use paypal, what then?

State departments will often purchase foreign goods, (say NZ government purchases Toyota EV cars from Japan for a parliamentary driver fleet). In these cases the purchase orders can be significant amounts, and so a tiny inflation pass-through effect could acrue — since the NZ government is normally not selling goods to get the foreign currency as a reserve balance at the foreign central bank.

The NZ government might not even have an account at a foreign central bank, and international payments sometimes could just go through commercial banks in the respective countries via SWIFT or these days even BRICS.

But is this source of inflation pass-through ever truly significant? Is it even measurable above forex trade noise levels? (A big driver of forex rates of change being the speculative trading itself, not “economic fundamentals.”)

A Few Answers

I do not get these answers from the RBNZ or FOIA requests, I can’t be bothered, these are just my “As It Could Be” Dirtbag MMT takes. I am far more interested in what can happen, not what does happen, and especially not what some brain-dead (aka. neoliberally indoctrinated) officials say happens. Still, i think operationally this is basically what does happen.

A PayPal user never directly “gets the foreign currency”; PayPal and the banking system intermediate via FX conversion at marked‑up wholesale rates, and at the macro level NZ’s import‑price pass‑through from exchange rates is measurable but generally modest and treated as transitory by the RBNZ.

Q1. PayPal FX mechanics

(a) Paying for foreign goods

PayPal gets wholesale FX quotes from its banking partners (essentially interbank rates) a few times per day and then puts a spread (typically around 3–4%) on top to create its retail rate.[4][5][1]

• When you pay a Singaporean or US seller from a NZD‑based account, PayPal debits your balance/card in NZD, converts at that internal rate, credits the seller in their chosen currency, and keeps the spread plus any cross‑border/merchant fee; you never hold SGD or USD yourself.[6][7][1]
• The “swap” here is not a central‑bank swap line but plain retail FX: PayPal (or its partner bank) is long one currency and short another, hedging its net exposure in the professional FX market while offering you a bundled payment+FX service.[1][4]

(b) Withdrawing to your domestic bank

• When you withdraw to a domestic bank account, PayPal will only send funds in the local currency of the account’s country; foreign‑currency balances are automatically converted to that local currency, again using PayPal’s own FX rate plus a conversion spread (often 3–4%).[5][8][1]
• If your PayPal balance is already in the same currency as your bank account, there is no FX conversion, only whatever withdrawal fee (e.g. instant‑transfer fee) applies; if currencies differ, PayPal does the FX before sending the transfer through normal banking rails.[8][1]
• In practice this means: a NZ user paid in USD on PayPal can either keep a USD “sub‑balance” and spend it on further USD‑denominated PayPal purchases, or convert to NZD inside PayPal (at PayPal’s rate) when withdrawing to a NZD bank account; in both cases, the FX leg remains within PayPal’s books and its banking counterparties.[9][1]

From an MMT lens: for the end user, both legs are just NZD transactions against a payments intermediary; the FX layer is a wholesale balance‑sheet translation between PayPal and its banks, not a retail acquisition of USD/SGD by the New Zealander.

As for the “PayPal Mafia” — they may such, but the payments layer itself is transparent, can be looked-up here [PayPal Consumer Fees]((https://www.paypal.com/us/digital-wallet/paypal-consumer-fees ) and is fairer than some services like Stripe and Patreon. I’ve used WISE in the past and they also have low fees.

The question you should ask is why the fees anyway? We could have pure public banking, zero fees, zero interest charges on loans. Why not? The answer is largely because of the neoclassical mindset of “Loanable Funds.” The government is not running a loanable funds system, they always have run MMT systems, since the first government in historical records.

Under the false neoclassical/New Keynesian framing the discipline on banks is on their liabilities (deposits). This is completely nuts. No typical customer can possibly have enough knowledge or nous to know how much of a risk their bank is at any time. Not even W. Buffet would know.

All despot holders should instead be insured by the State, so if a bank is closed (by the Statutory authorities) no deposit holders lose their deposits, they just get transferred in whole to another bank.

The MMT Case is that the discipline on banks has to be on the Asset side. Governments should be telling banks only what they can do — and they cannot do anything else. Fore example: (1) Clear payments, (2) Assess credit- worthiness, then go home for the day.

Q2. Government imports and inflation pass‑through

How the government actually pays

When the NZ government buys, say, Toyota cars from Japan, it instructs its bank (or the NZDM’s banking counterparties) to make a foreign payment; the NZ bank debits NZD from the Crown account at the RBNZ, obtains JPY in the FX market (directly or via its own correspondent bank), and credits the Japanese exporter’s bank in JPY over SWIFT (or other messaging networks).

The NZ government itself does not need an account at the Bank of Japan; NZD liabilities are created by domestic fiscal operations and then swapped for foreign currency via commercial banks and FX markets, just as for large private‑sector importers, but at larger ticket sizes and often on better terms.[10][3]

NB: There is an RBNZ model for forex pass-through, but would you trust them with your savings, or touch their models with a barge pole? This article by Sherwin (2000) talks about inflation targeting, which from an MMT standpoint is weak, since the proper target is full employment — provide you desire to optimize use of real resources available. ((Maybe they don’t?)

This paper Tiuca ＆ Posedel (2009) is useful because it has an empirical model, albeit only for a case study in Croatia. It shows during low forex-to-price pass-through effect regimes the domestic prices adjust rapidly, that is, adjustments in relative trade balances is fairly prompt, that is, the **real exchange rate ** (relative value of goods) changes rapidly in response to a forex change.

On the other hand, if the regime is one of high forex-to-price pass-through effect then the nominal exchange rate does not have much impact on the real exchange rate (prices of goods domestically change little relative to foreign goods).

Not sure about you, but to me this implies we should desire high pass-through effect. This is “in the macro”. Obviously for import-business and export-business they are always either winners or losers. The domestic macroeconomy should not care too much about that rebalancing. We should want greater purchasing power parity — hence rapid response and no significant change in the real exchange rate. ((Sure, I understand Imperialists are always after something else, something pirate-like and screwy to their donor-firm’s benefit.))

Another model is this one from [Ha et al (2020)]((https://papers.ssrn.com/sol3/Delivery.cfm?abstractid=3360138) . Lo and behold! “Pass-through ratios tend to be lower in countries that combine flexible exchange rate regimes and credible inflation targets.” — who could’ve 𝕗𝕣𝕚𝕘𝕘𝕚𝕟 guessed? (Not “MMT” since they’re a loony cult, right?)

Then this article by Takhtamanova (2008) . Low and behold! Low inflation rate (low interest rate floor) era implied less forex pass-through!! Rocket Science!!!

Let me know if any of these is empirically useful! I’m not paying them any attention, they are built upon false assumptions.

Magnitude of exchange‑rate pass‑through for NZ

Empirical work on New Zealand shows that exchange‑rate pass‑through to import prices is partial at the level of individual goods but close to complete for aggregate import prices over time; however, the impact on CPI is damped, delayed, and far from one‑for‑one. (Citing here [Heagney (1998)])(https://mro.massey.ac.nz/server/api/core/bitstreams/092d5586-f7e2-45ad-930c-e85000583ae1/content ) – maybe a touch out of date? Lo and behoild! This pass-through data shows New Zealand (rapid pass-though) is in fact a “small country” —- who would’ve guessed?

A detailed study of NZ’s major imports (1988–1997) finds partial first‑quarter responses and incomplete pass‑through for specific products, while aggregated imports show faster, more complete pass‑through; even then, transmission from import prices to CPI is diluted by distribution margins, wages, and non‑tradable components.

More recent RBNZ practice explicitly treats the direct price effects of exchange‑rate swings as largely transitory “noise” and does not automatically offset them with interest‑rate changes, on the assumption that such shocks wash out unless they trigger second‑round effects in wages and expectations.

Is the government’s own FX demand an important driver?

In a small open economy such as NZ, total FX turnover is dominated by private trade, portfolio flows, and speculative/hedging activity; government goods imports are a tiny fraction of daily FX market volume and are typically executed via banks that net flows and hedge in deep markets.

As a result, the incremental FX demand associated with the government’s import program is not large enough to be reliably separable from “forex noise” generated by private flows and speculation; empirically, studies discuss pass‑through from exchange‑rate movements in general, not from official‑sector purchases as a distinct driver.

The inflation‑relevant variable is the exchange rate itself and how much of its movement passes through into domestic prices, not who triggered marginal trades; central banks model pass‑through from the observed nominal exchange rate path and usually treat government purchases as macro‑trivially small relative to market depth.

From an MMT perspective: since the state, especially on a floating exchange rate, does not run out of foreign currency in a mechanical sense; the binding constraint is still the real terms at which it can exchange its own liabilities for foreign goods and assets. That constraint shows up as shifts in the exchange rate and the associated (modest, empirically partial) import‑price pass‑through, not as a discrete, independently measurable government FX‑import inflation channel separate from the broader exchange‑rate process.

Why would we say “especially on a floating exchange rate”?

Well, that’s just Repolitik. On a fixed exchange rate regime NZ can still swap as many NZD for any foreign currency needed to clear payments, but with weak exports may suffer politically undesirable domestic inflation, at least one-off upwards price adjustment. The voting electorate is badly educated, and prefer to see unemployment rather than some price adjustments, even when their wages are going up to match! It’s an horrific mental disease.

But this is onyl a psychological constraint, it is not a monetary operations constraint. THe NZ government can still, always and at all times, ensure full non- 🅱𝘂𝓵🅛🅂𝖍𝗶🅃 job employment, and enjoy a decent standard of living, free from precarity for all. Why don’t we?