UK Investment Report 2016 — Pricing an Unresolvable Question

The UK's June 2016 referendum result created an unusual analytical problem: a question that would take years to resolve and had to be priced immediately.

What the withdrawal would mean in practice — for trade, for the movement of people, for financial services access, for regulatory alignment — depended on negotiations that had not begun. Markets could not wait for the answer. They had to price a probability distribution and revise it as information arrived.

The adjustment concentrated in the currency, and understanding why is the most useful part of the episode.

Sterling fell substantially and did not recover on the timeline that equity indices did. The equity index, by contrast, fell briefly and then rose — in a way frequently misremembered as markets shrugging off the result.

They did not shrug it off. The adjustment happened somewhere else.

A large share of the earnings of major UK-listed companies is generated abroad and denominated in foreign currency. When sterling falls, those earnings translate into more sterling. The index rises in sterling terms without anything improving — and for a foreign investor holding unhedged, the sterling gain is offset by the currency loss.

The index and the economy are different things, and 2016 separated them unusually clearly. Beneath the index, the split was stark: internationally-oriented companies rose, domestically-oriented ones — retailers, housebuilders, domestic banks — fell.

Separately, a structural problem surfaced immediately. Several open-ended property funds suspended redemptions within days, demonstrating a liability mismatch that recurs whenever daily-dealing vehicles hold assets that take months to sell.

Why a currency absorbs an unresolvable question

The concentration of the adjustment in sterling was not arbitrary. It reflects what a currency is and what an equity index is.

A currency is a claim on an economy as a whole. It reflects the aggregate assessment of that economy's prospects, its trade position, its policy path and its attractiveness to capital. It can move continuously and by any amount, and it does not require a view about any particular company.

An equity index is a claim on a specific set of companies, weighted by size. Those companies have their own characteristics — where they earn, what they sell, who their customers are — that may or may not track the economy in which they are listed.

When a question arises about an economy's prospects, the currency is the instrument best suited to expressing it, for three reasons:

• It is the aggregate exposure. A view about the economy is naturally expressed in the currency.
• It is liquid and continuously traded, so a view can be expressed at any scale.
• It requires no company-level view. An investor uncertain about the economy but not about any specific business can act.

A currency can price a question about a country. An equity index can only price a question about the companies in it — and in an open economy, those are frequently not the same question.

The practical implication for any investor in an open economy is that a country-level shock will express itself in the currency first, and that the equity index's response depends on the index's composition rather than on the shock's severity. An index dominated by international earners is a poor instrument for expressing a domestic view, and using it as one produces exactly the confusion 2016 produced.

The composition split

The divergence beneath the index level is the clearest illustration of the point, and it is worth setting out because the same structure appears in every open economy.

Companies that rose shared a characteristic: foreign earnings. Large multinationals in resources, pharmaceuticals, consumer goods and industrials earn substantially abroad. A weaker home currency raises their reported earnings mechanically and improves the competitiveness of anything they export.

Companies that fell were domestically oriented: retailers, housebuilders, domestic-focused banks, real estate, consumer services. These face the opposite effects — costs of imported goods rise, domestic demand is uncertain, and there is no foreign earnings offset.

The dividing line was where the earnings come from, not any judgement about the referendum outcome.

Two second-order effects are worth noting:

• Import costs rose for domestic businesses. A retailer sourcing goods abroad faces higher costs in sterling, and passing them on depends on pricing power. Margins compress where it is absent.
• Inflation rose, since imported goods cost more. This affects household real incomes and constrains monetary policy — a weaker currency raising inflation limits the central bank's ability to cut rates in response to weaker growth.

That last point is the awkward one and it recurs in every currency-driven adjustment. A currency fall provides a competitiveness benefit and an inflation cost simultaneously, and the central bank facing weaker growth alongside higher inflation has no clean response. This is the same bind that faced multiple economies in 2022, and it is why currency adjustment is a real adjustment rather than a free absorber.

Property funds and the liability mismatch

Within days of the referendum, several open-ended property funds suspended redemptions. The episode is a clean demonstration of a structural problem that recurs.

The structure. An open-ended fund offers daily or near-daily redemption. The fund holds commercial property, which takes months to sell at a reasonable price.

The mismatch is obvious when stated and is accepted because it works most of the time. In normal conditions, redemptions are modest and are met from cash buffers or from the fund's ongoing inflows. The mismatch is invisible.

In stress, three things happen together:

• Redemption requests rise, because investors seek liquidity precisely when they are most concerned.
• The buffer is consumed, and replenishing it requires selling property, which takes months.
• Selling into a stressed market realises poor prices, disadvantaging the investors who remain.

Suspension is the mechanism that resolves this, and it protects remaining investors from having assets sold at distressed prices to fund exits. It also means investors who wanted liquidity cannot have it — which is precisely what the fund had promised.

There is a further consideration. The fund's net asset value is based on property valuations that are appraisal-based, per the mechanism the 2015 private equity report describes. An investor redeeming receives that NAV. If appraisals are above realisable value in a falling market — which is the standing concern — early redeemers receive more than the assets are worth and the cost falls on those who remain. The valuation question and the liquidity question are the same question, exactly as in the semi-liquid private credit vehicles the 2024 report describes.

The general principle: a fund's liquidity terms should match its assets' liquidity. Where they do not, the fund is offering something it cannot always deliver, and the promise fails precisely when it is most valued. This recurs — in 2008, in 2016, in 2020, and in every subsequent episode where daily-dealing vehicles held illiquid assets.

The cost of prolonged uncertainty

The market adjustment resolved within months. The uncertainty did not, and its investment cost was in a different form.

The negotiation extended for years, with the eventual arrangements unclear throughout. During that period:

• Investment decisions were deferred. A company deciding whether to build in the UK, or to relocate a function elsewhere, faced an unanswerable question about the future regulatory and trading environment. As the 2018 Asia-Pacific report describes for trade policy, the rational response to unresolved uncertainty about an irreversible decision is frequently to wait — and the cost of waiting is incurred regardless of how the question resolves.
• Financial services faced a specific question about access to European markets, prompting contingency arrangements — establishing subsidiaries, relocating staff, obtaining licences — that cost money whether or not they proved necessary.
• Labour market effects developed gradually, with implications for sectors dependent on workers from elsewhere in Europe.
• Regulatory divergence became a possibility, requiring businesses to plan for two regimes.

The measurable cost was in deferred investment, not in the market move, and it accrued over years rather than in days. This is the general pattern with political and regulatory uncertainty: the market impact is immediate, visible and often overstated; the economic impact is gradual, diffuse and usually larger.

For investors, the durable lesson is that the market's reaction to a political event and the event's economic consequences operate on different timescales, and that observing a market recovery is not evidence that the consequences were small.

The market's return and the investor's return

2016 separated two numbers that are routinely treated as one, and the separation is worth generalising because it applies to every cross-border holding.

The market's return is what the assets did in their own currency. This is what index providers report and what commentary discusses.

The investor's return is what they experienced in their own currency. It is the market's return combined with the currency move, and for an unhedged holder the two can differ enormously.

In 2016 the divergence was extreme in both directions:

• A sterling-based holder of UK equities saw the index rise. Their return was positive.
• A dollar-based unhedged holder of the same index saw the index rise in sterling and sterling fall against the dollar. Their return was substantially worse and in several periods negative.
• A sterling-based holder of foreign assets saw the reverse: foreign holdings translated into more sterling, producing gains that had nothing to do with the assets.

The mechanism has a specific and awkward property. In an open economy, the currency and the equity index are frequently negatively correlated — because a weaker currency raises the translated foreign earnings that dominate the index. That negative correlation means an unhedged foreign investor experiences a partially self-cancelling position, in which the index's rise is offset by the currency's fall. They are, in effect, hedged by accident and imperfectly.

The consequences for how a foreign allocation should be constructed:

• Hedging is a decision even when it is not made. An unhedged position is a currency position of the same size as the equity position, taken by default.
• The hedge changes the exposure, not only the volatility. Hedging a UK equity position converts it from a partially self-cancelling position into a direct claim on those companies' earnings. That is a different exposure and it may or may not be the intended one.
• The composition of the index determines which is appropriate. An index of international earners behaves differently from an index of domestic ones, and the FTSE 100 versus FTSE 250 divergence in 2016 is the clearest available illustration.

"UK equities rose in 2016" and "foreign investors in UK equities did well in 2016" are both statements about the same assets. Only one of them is true, and which one depends entirely on a decision most investors did not consciously make.

What an allocator could act on

Set hedging policy in advance. After a currency has moved, the decision is no longer available. This is the single most consequential and most frequently defaulted decision in cross-border allocation.

Look at the index's earnings geography before treating it as a country exposure. An index dominated by companies earning abroad is not a claim on the domestic economy. If the intended exposure is domestic, a mid-cap or domestically-oriented index is the instrument, and the difference between the two was very large in 2016.

Match a fund's redemption terms to its assets' liquidity. The property fund suspensions demonstrated that a daily-dealing vehicle holding assets that take months to sell is offering something it cannot always deliver. The promise fails precisely when it is most valued. This is checkable from the fund's own documents before investing, and it recurs — in 2008, 2016, 2020, and in the semi-liquid private credit vehicles the 2024 report describes.

Recognise that redemption at NAV in an appraisal-valued fund transfers value. If the appraisal is above realisable value in a falling market, early redeemers receive more than the assets are worth and the cost falls on those who remain. The valuation question and the liquidity question are the same question in any vehicle offering redemption at a modelled price.

Separate the market timeline from the policy timeline. The market adjustment resolved in months; the negotiation ran for years. During that interval the economic cost accrued in deferred investment, which does not appear in a price series. Observing a market recovery is not evidence that the consequences were small — it is evidence that the market had finished repricing, which is a different statement.

Track the uncertainty, not just the outcome. Trade and regulatory uncertainty indices are published free and they measure the variable that was actually suppressing investment. They move independently of the eventual policy result, which is precisely why they are informative: the cost was being incurred while the question stayed open, and it was measurable throughout.

What 2016 established for UK analysis

• The currency absorbed the adjustment, because it is the instrument suited to an aggregate question.
• The index and the economy were separated clearly, with foreign earnings composition determining the index's response.
• Sector composition was the dividing line, not any judgement about the outcome.
• The property fund episode demonstrated the liability mismatch and the identity of the valuation and liquidity questions.
• The prolonged uncertainty's cost was in deferred decisions, accruing over years and larger than the market move.