What does Cross-Border Settlement Friction and Conversion Batches conclude, in numbers, about “A Network of Currency Channels: Minimum-Loss Flows and the Shadow Price of Money”?
Answer: Chapter 3 draws a group's currency accounts and licensed channels as a network and finds by minimum-cost flow a route losing 25.69 basis points against 28.15 for direct return, with account shadow prices and bargaining targets for channels.
Scope of the numbers: All fees, rates, volatilities, amounts and group structures are illustrative; all calculations are generated by the book's programs; the book is not individual legal, tax or financial advice. Marked boxes separate published evidence, runnable Python calculations (19 programs), classical sources and the author's hypotheses.
Data table
The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 25.69, 28.15
| minimum-loss plan (kUSD-equivalent per month) | ||||
| US.USD | > HK.USD | wire | 6 bp | 100 |
| US.USD | > CN.USD | wire | 10 bp | 300 |
| DE.EUR | > HK.EUR | wire | 5 bp | 250 |
| HK.EUR | > HK.USD | FX EUR>USD | 12 bp | 250 |
| HK.USD | > CN.USD | wire | 4 bp | 150 |
| HK.USD | > HK.CNH | FX USD>CNH | 9 bp | 150 |
| HK.CNH | > CN.CNY | RMB cross-border | 6 bp | 150 |
| CN.USD | > CN.CNY | FX tier 1 | 10 bp | 300 |
| CN.USD | > CN.CNY | FX tier 2 | 20 bp | 150 |
| optimal loss 1670.0 USD = 25.69 bp of 650 kUSD | ||||
| direct-to-home loss 1830.0 USD = 28.15 bp | ||||
| saving 160.0 USD a month, 0.087 of the direct loss |
Show the full program
# Chapter 3. p31: minimum-loss routing over a multi-currency channel net.
# A group owes its home entity (CN, needs CNY) for goods under real
# intra-group sales contracts; the US and DE entities hold the receipts.
# Nodes are (entity, currency) accounts; arcs are licensed channels:
# same-currency transfers and same-entity conversions. Every arc costs
# whole basis points of the amount and some have monthly capacities.
# Amounts are kUSD-equivalent at mid rates; 1 bp on 1 kUSD = 0.1 USD.
# Successive shortest paths with Dijkstra on reduced costs (heapq),
# integer arithmetic throughout. All fees are illustrative only.
import heapq
NODES = ["US.USD", "DE.EUR", "HK.USD", "HK.EUR", "HK.CNH",
"CN.USD", "CN.EUR", "CN.CNY"]
BIG = 10 ** 6
ARCS = [ # from, to, capacity kUSD, bp, channel
("US.USD", "HK.USD", BIG, 6, "wire"),
("US.USD", "CN.USD", BIG, 10, "wire"),
("US.USD", "CN.CNY", 200, 35, "PSP collect+FX"),
("DE.EUR", "HK.EUR", BIG, 5, "wire"),
("DE.EUR", "CN.EUR", BIG, 12, "wire"),
("HK.EUR", "HK.USD", BIG, 12, "FX EUR>USD"),
("HK.USD", "CN.USD", BIG, 4, "wire"),
("HK.USD", "HK.CNH", 150, 9, "FX USD>CNH"),
("HK.CNH", "CN.CNY", BIG, 6, "RMB cross-border"),
("CN.USD", "CN.CNY", 300, 10, "FX tier 1"),
("CN.USD", "CN.CNY", BIG, 20, "FX tier 2"),
("CN.EUR", "CN.CNY", BIG, 30, "FX EUR>CNY")]
SUPPLY = {"US.USD": 400, "DE.EUR": 250, "HK.USD": -50, "CN.CNY": -600}
def min_cost_flow(nodes, arcs, supply):
idx = {v: k for k, v in enumerate(nodes)}
n = len(nodes) + 2
src, snk = n - 2, n - 1
g = [[] for _ in range(n)]
def add(u, v, cap, cost):
g[u].append([v, cap, cost, len(g[v])])
g[v].append([u, 0, -cost, len(g[u]) - 1])
return (u, len(g[u]) - 1)
handles = [add(idx[a], idx[b], c, w) for a, b, c, w, _ in arcs]
need = 0
for v, s in supply.items():
if s > 0:
add(src, idx[v], s, 0)
need += s # step
elif s < 0:
add(idx[v], snk, -s, 0)
pot = [0] * n
cost = flow = 0
while flow < need:
dist = [None] * n
prev = [None] * n
dist[src] = 0
heap = [(0, src)]
while heap:
d, u = heapq.heappop(heap)
if d != dist[u]:
continue
for k, (v, cap, w, _) in enumerate(g[u]):
nd = d + w + pot[u] - pot[v]
if cap > 0 and (dist[v] is None or nd < dist[v]):
dist[v] = nd
prev[v] = (u, k)
heapq.heappush(heap, (nd, v))
if dist[snk] is None:
raise ValueError("infeasible")
for v in range(n):
if dist[v] is not None:
pot[v] += dist[v] # step
push, v = need - flow, snk
while v != src:
u, k = prev[v]
push = min(push, g[u][k][1])
v = u
v = snk
while v != src:
u, k = prev[v]
g[u][k][1] -= push # step
g[v][g[u][k][3]][1] += push # step
cost += push * g[u][k][2] # step
v = u
flow += push # step
used = [arcs[j][2] - g[u][k][1] for j, (u, k) in enumerate(handles)]
return cost, used, g, handles
if __name__ == "__main__":
cost, used, _, _ = min_cost_flow(NODES, ARCS, SUPPLY)
print("minimum-loss plan (kUSD-equivalent per month)")
for (a, b, cap, bp, ch), f in zip(ARCS, used):
if f:
print(" %-7s > %-7s %-17s %3d bp %5d" % (a, b, ch, bp, f))
naive = [("US.USD", "CN.USD", 10, 350), ("CN.USD", "CN.CNY", 10, 300),
("CN.USD", "CN.CNY", 20, 50), ("DE.EUR", "CN.EUR", 12, 250),
("CN.EUR", "CN.CNY", 30, 250), ("US.USD", "HK.USD", 6, 50)]
naive_cost = 0
for a, b, bp, f in naive:
naive_cost += bp * f # step
vol = 650
print("optimal loss %.1f USD = %.2f bp of %d kUSD"
% (cost / 10.0, cost / vol, vol))
print("direct-to-home loss %.1f USD = %.2f bp"
% (naive_cost / 10.0, naive_cost / vol))
print("saving %.1f USD a month, %.3f of the direct loss"
% ((naive_cost - cost) / 10.0, 1 - cost / naive_cost))Single source and citation
Fan, Y. (2026). Cross-Border Settlement Frictions and Compliant Treasury Scheduling: Minimum-Loss Network Flows over Multi-Currency Channel Fees and Conversion Slippage (Version 1.0.0), Chapter 3: A Network of Currency Channels: Minimum-Loss Flows and the Shadow Price of Money. 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-280 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026280,
author = {Fan, Yafu},
title = {Cross-Border Settlement Frictions and Compliant Treasury Scheduling: Minimum-Loss Network Flows over Multi-Currency Channel Fees and Conversion Slippage},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 1 · Each Receipt or in Batches: One Month's Ledger of Conversion Friction
- Chapter 2 · The Price of Waiting: How Exchange-Rate Risk Pushes the Square Root towards a Cube Root
- Chapter 4 · Offset Before You Transfer: Multilateral Netting and the Fewest Payments
- Chapter 5 · A Month's Conversion Calendar: Dynamic Programming, Minimum Charges and Volume Tiers
- Chapter 6 · Robustness: Wrong Parameters, Fat-Tailed Rates and a Written Loss Budget
- Chapter 7 · Eastern and Western Sources Examined: the Quanfu Office Gathers and Releases, Fibonacci Chains the Rates
- Chapter 8 · Closing the Loop: the Monthly Batch Table, the Conversion Batch Calculator and a Pre-Registered Test