Cross-Border Inflation Defence for Family Offices

Three statistical agencies published an inflation number for July 2026. Statistics Canada said 3.0%. The Bureau of Labor Statistics said 3.4%. Taiwan’s Directorate General of Budget, Accounting and Statistics said 2.54%. A family with a house in Vancouver, an apartment in Taipei, and a portfolio denominated in US dollars received three answers to a question it had asked once, and none of the three described what happened to its purchasing power.

Cross-border inflation defence begins with a measurement problem before it becomes an allocation problem. The rate that matters to a family living across jurisdictions is a blend, weighted by where the family spends and translated at the rate where the family holds. No agency publishes that index, because no agency has the family’s balance sheet. Building it is step one. Step two is choosing hedges that survive contact with three tax regimes, which is where most of the damage is done, because the instrument that hedges inflation efficiently in one jurisdiction is frequently taxed, situs-trapped, or structurally unavailable in another.

Three Inflation Rates, One Balance Sheet

The July 2026 readings sit close enough together to look interchangeable.

A spread of under a percentage point across the headline numbers is not what moves a cross-border balance sheet. Currency is. The Canadian dollar traded near 1.4793 against the US dollar in February 2025 and around 1.38 in August 2026, a move of roughly seven percent. The New Taiwan dollar has ranged between about 31.1 and 32.2 to the US dollar within 2026 alone, after a rally in 2025 that reversed. Those translation effects dwarf the CPI differentials several times over, and they arrive without warning attached to a central bank statement.

The practical consequence is that a family’s real return target has to be stated against a liability, in the currency of that liability. Tuition payable in US dollars, property tax payable in Canadian dollars, and household costs payable in New Taiwan dollars are three separate obligations with three separate inflation paths. Aggregating them into one number and comparing that number to a portfolio return produces a figure that feels rigorous and measures nothing.

Building the blend is unglamorous work. It requires the family to state, in writing, what proportion of annual spending falls in each currency, and to revisit that split when a household relocates or a next-generation member starts a career abroad. The multi-generational investment strategy framework that governs the portfolio should carry this weighting as an input, not as a footnote.

The Hedge Canada Stopped Selling

On 3 November 2022, the federal government ceased issuing Real Return Bonds, the Government of Canada instruments whose principal is indexed to the CPI. They had been issued regularly since 1991. Existing bonds continue to trade in the secondary market until maturity, some as far out as 2044, but no new supply is entering. The last provincial issuance dates to 2008, and the Canadian corporate inflation-linked market is effectively nonexistent.

For a Canadian-resident family office, this removes the domestic instrument that would otherwise be the default answer to inflation-linked liabilities. What remains is a shrinking, thinly traded pool where holders are reluctant sellers and price discovery is poor. The Canadian Institute of Actuaries and the C.D. Howe Institute have both argued publicly for reversing the decision. That has not happened.

So Canadian families reach for US Treasury Inflation-Protected Securities instead, which is a reasonable substitute in economic terms. It also converts a domestic inflation hedge into a US dollar position held against Canadian dollar liabilities. The hedge now carries a currency bet inside it, and the family has quietly increased the exposure it was trying to reduce. Whether that trade is worth making depends on the currency split of the liabilities it funds, which returns the exercise to step one.

Where the Wrapper Decides the Outcome

Here the cross-border problem stops being a portfolio question and becomes a structuring question, and the difference is worth real money.

A Canadian or Taiwanese resident who is neither a US citizen nor a US domiciliary is exposed to US federal estate tax on US-situs assets above US$60,000, at rates reaching 40%. Direct holdings of US Treasury securities, including TIPS, generally fall outside that net under the statutory carve-outs for US government obligations and portfolio debt. Shares of a US-domiciled fund do not. A US-listed TIPS ETF is stock in a US entity, and it sits squarely inside the situs rules.

The same underlying exposure, purchased through the convenient wrapper rather than the direct instrument, changes the estate outcome entirely. The same logic reaches the real asset sleeve: US REITs are shares of US corporations, so the property allocation most commonly used as an inflation hedge is US-situs by construction. Funds domiciled outside the United States that hold identical underlying assets are generally not.

None of this is exotic. It is a wrapper decision that gets delegated to whoever executes the trade, made on the basis of expense ratio and liquidity, and never reviewed against the family’s estate exposure. The mechanics of situs, treaty proration, and the marital credit belong in a dedicated review of US estate tax reach rather than in an allocation memo, but the decision that triggers them is made at the trading desk.

The PPLI Table That Does Not Travel

The most widely circulated table in family office media on this subject shows a private credit strategy yielding 10% gross netting down to roughly 4.6% for a taxable investor in a high-tax US state, against roughly 9.4% for the same strategy held inside a Private Placement Life Insurance policy. The arithmetic is correct. It also assumes a US-taxable policyholder holding a policy compliant with section 7702 of the Internal Revenue Code, and it is shown to cross-border families as though those assumptions were incidental.

They carry the whole comparison. Under Canadian rules, a foreign life insurance policy owned by someone who becomes a Canadian resident is treated as last acquired at the point of residency. From then on it is subject to annual accrual taxation under subsection 12.2(1) of the Income Tax Act on the amount by which its accumulating fund exceeds its adjusted cost basis, unless the policy qualifies as an exempt policy under section 306 of the Income Tax Regulations.

The exempt test asks, in substance, whether a policy is primarily designed to fund a death benefit or primarily designed as an investment and savings vehicle. PPLI is engineered as the second thing. Policies issued by insurers outside Canada are not excluded from qualifying, and the CRA reconfirmed at the May 2026 CALU roundtable that exempt testing has to be performed at each policy anniversary regardless of any contractual guarantee the insurer has given. But the testing has to actually be done, by an actuary familiar with the Canadian regulations, and the answer is not knowable from the marketing material.

There is a further wrinkle that connects directly back to the currency problem. The CRA has acknowledged that where a policy is not denominated in Canadian dollars, applying the exempt test can produce accidental failure purely from year-to-year movement in the conversion rate. The currency exposure a family took on while reaching for a US-based inflation hedge can, in that scenario, break the tax wrapper that was supposed to protect the return the hedge produced.

This is licensed terrain and it should be treated that way. Determining exempt status requires an actuary; determining the Canadian and US tax consequences of a specific policy requires cross-border tax counsel and, where a trust holds the policy, the lawyer who drafted it. What belongs at the family office level is the question that triggers the referral, which is whether anyone has tested the assumptions in the table before the family relied on it.

Three Thresholds, Three Indexation Rules

Tax thresholds inflate at different speeds too, and the differences compound across a generation.

Taiwan’s step-function approach means the exemption erodes in real terms between adjustments. A Taiwan-resident family member whose household overseas income reaches NT$1 million in a year has that income pulled into the basic income calculation, and the threshold shielding it is worth progressively less each year until the CPI trigger is finally hit. The US branch of the same family, holding an equivalent position, sits behind a threshold that moves every January. The Canada Taiwan arrangement and its withholding and residency mechanics determine which branch is in which position.

Canada’s treatment deserves separate attention because it is the least visible. Capital gains are computed on nominal amounts. A property bought twenty years ago and sold today is taxed on the full nominal gain, including the portion that represents nothing more than currency debasement. Inflation is therefore taxed directly, at the marginal rate, on realization. Foreign currency gains are treated the same way under subsection 39(1.1), with a de minimis of only $200 a year for individuals, which means the currency hedge itself generates a taxable event when it works.

One Family, Several Real Return Targets

The governance conclusion follows from the arithmetic. An investment policy statement specifying a single real return objective, expressed as CPI plus a spread, cannot serve a family whose members are resident in different jurisdictions. Each branch faces a different consumption currency, a different marginal rate, and a different indexation regime. One target number papers over three different problems.

The workable version states the return objective per liability pool, in the currency of that pool, net of the tax that applies where the funding assets are held. It also specifies which entity holds which hedge, because the situs and exempt-test questions above are determined by ownership rather than by allocation. Reworking an investment policy statement to carry that structure is a governance exercise, and it is far cheaper to do before an estate settles than after.

Two further pieces belong in the same review. Distribution policy should specify the currency of each distribution, so that a payout funded in one currency and consumed in another does not quietly transfer FX risk to a household that has no capacity to manage it. And the private market sleeve, where CPI-linked concession revenue and other contractual inflation linkages actually live, should be assessed for what the allocation to real assets and private credit is doing net of the wrapper and holding-entity decisions rather than gross. Families already thinking about jurisdictional exposure in geopolitical terms usually find the inflation analysis lands on the same map.

Frequently Asked Questions

Does PPLI work for a Canadian resident?

It can, but not on the assumptions used in US marketing material. A policy issued outside Canada and owned by a Canadian resident is deemed last acquired on the date residency begins, and is then subject to annual accrual taxation under subsection 12.2(1) of the Income Tax Act unless it qualifies as an exempt policy under section 306 of the Income Tax Regulations. The exempt test distinguishes policies primarily designed to fund a death benefit from those designed as investment vehicles, which is the profile PPLI deliberately adopts.

Foreign-issued policies are not excluded from qualifying, and the CRA confirmed in May 2026 that exempt testing must occur at every policy anniversary. Establishing whether a specific policy qualifies requires an actuary familiar with the Canadian regulations, working alongside cross-border tax counsel. The net return figures published for US taxable investors should not be applied to a Canadian-resident policyholder without that work.

What replaced Real Return Bonds for Canadian investors?

Nothing domestic. The federal government stopped issuing Real Return Bonds on 3 November 2022. Existing bonds continue trading until maturity, with some running to 2044, but supply is fixed and liquidity has thinned.

Provincial issuance ended in 2008 and there is no meaningful Canadian corporate inflation-linked market. Canadian investors seeking contractual inflation linkage now use US TIPS, inflation-linked bonds from other sovereigns, or private real assets with CPI-indexed revenue under concession or regulatory frameworks. Each substitution introduces something the RRB did not carry, usually currency exposure, and sometimes US situs exposure depending on how the position is held.

Do US TIPS create US estate tax exposure for a Canadian resident?

Held directly, generally no. US government obligations fall within the statutory carve-outs that keep them outside the definition of US-situs property for a non-resident, non-citizen decedent.

Held through a US-domiciled ETF or mutual fund, the answer changes: those fund shares are stock in a US entity and are US-situs, exposing the estate to federal estate tax above the US$60,000 threshold at rates up to 40%. The same distinction applies to US REITs, which are US corporations regardless of the underlying property. Funds domiciled outside the United States holding equivalent assets are generally not US-situs. The wrapper decision, often made on cost and convenience grounds, determines the estate outcome.

How should a cross-border family set its real return target?

Per liability pool, in the currency of the liability, and net of tax where the funding assets are held. A family with obligations in Canadian dollars, US dollars, and New Taiwan dollars has three targets rather than one. Setting a single objective against a single CPI series produces a number that cannot be checked against anything the family actually pays for. The investment policy statement should also record which entity holds each position, since situs treatment and insurance exempt-test status are determined by ownership.

How does Taiwan tax a resident family’s overseas investment income?

Through the Income Basic Tax rather than the regular income tax. Where a tax household’s aggregate overseas income reaches NT$1 million in a year, that income is included in the basic income calculation. Basic income above the NT$7.5 million exemption is taxed at a flat 20%, and the taxpayer pays the difference where the basic tax exceeds regular tax. The exemption amount only adjusts when cumulative CPI growth reaches 10%, so it held flat for 2026 and continues to lose real value between adjustments. Overseas income is not included in the reference data the tax bureau provides, so the reporting obligation sits entirely with the family.

Most of the inflation literature aimed at family offices is written for a reader who lives, spends, invests, and dies in one country, and it is perfectly good advice for that reader. For a family whose balance sheet crosses the Pacific, the same advice can produce an inflation hedge that is fully effective in economic terms and quietly expensive in tax terms, which is a harder failure to spot because the portfolio report shows it working. Testing which hedges survive translation is a discrete piece of work, best done before the next rebalance rather than during an estate administration, and it is a natural first project for a family that has not yet built the coordinating layer described in the family office setup guide. If your holdings, residences, and liabilities do not all sit in the same currency, that is a conversation worth having properly.