One Takeaway, Four Implications, and Seven Denials at 2026 IMF Annual Meetings

The team just returned from 2026 Annual IMF Meetings in Bangkok, where we discussed all key topics with key global policymakers, investors, and analysts. We’ve been attending these meetings for over three decades, so our attitude will be meta. Your author left Thailand at age 13 in 1979, and makes some additional observations on what has changed in Thailand since then. We appreciate your attention — there are a lot of details on individual countries inside, and they’re important, but we try always to start with a high-level view; not everyone wants to hear about the latest tilt in Chilean monetary policy at the beginning, or ever, and you don’t have to.

• “Fiscal dominance” is now the dominant frame for looking at all sovereigns. We have used that framing since 2012; G-10 colleagues who dismissed it as unlikely in DMs now acknowledge it.
• The US is exceptional again. Relative to the major DMs (EZ, Japan, UK) the US is in far better shape, and the structure of the Eurodollar system means money flows from those markets to the US.
• Europe is extremely vulnerable for well-known reasons. With no fiscal and financial union, individual sovereign bond markets such as France and Spain invite attack, and France itself will have to fix France.
• EMs remain largely exceptional, for durable reasons, but EM is not a monolith. Low government debt, less levered financial institutions and independent central banks support them; China and Thailand are poster-children for reserve-quality currencies.
• EMs are not all following the Fed. Mexico had already diluted that principle, and India and Indonesia will be tested; the divergence comes from strength (anchored inflation, low debt and leverage).
• DM political risk. Markets still assume political stability, but even in the US the K-shaped recovery is becoming an intractable political challenge. Asset conclusion: higher bond yields.
• US exceptionalism. The realization that the US is “least-bad” among DMs is just dawning, and Europe no longer gets the benefit of the doubt. Asset conclusion: stronger US dollar.
• Everyone remains an expert on geopolitics. Participants largely ignore major shifts (the Gulf closure, the Black Sea blockade) and the propaganda in each side’s narrative. Asset conclusion: higher-for-longer oil.
• China was everywhere and nowhere. Participants had little exposure, but we think the CNY will appreciate while external surpluses continue indefinitely. Asset conclusion: gradually stronger CNY and CGB prices.
• India drifting, at best. Already heterodox, with repeated INR intervention, no obvious bullish catalysts and real rates too low to anchor; INR is a zero-weight in our strategies. Asset conclusion: zero weight IGBs.
• EM is not a monolith — there are winners as well as losers. Winners include Brazil, Colombia, Malaysia, Thailand and China; losers include Indonesia, India, Philippines, Turkey and Mexico. Asset conclusion: o/w winners, u/w losers.
• AI is only about capex and productivity. That lens no longer works: AI debt issuance in 2026 transcends sovereign issuance, and rising rates will find problems. Asset conclusion: UST prices, short term, depend on AI debt issuance.

The Top Takeaway (and Its 4 Asset Price Implications)

“Fiscal dominance” is now the dominant frame for looking at all sovereigns. This was a long time coming, with ourselves using that framing since our first pieces in 2012. Your author has recounted his endless (happy) debates with his G-10 colleagues while at Morgan Stanley, who generally dismissed “fiscal dominance” as extremely unlikely in DMs. Not anymore.

The asset-price implications of this, broadly (details on key countries inside), are:

1. The US is exceptional again. We’ve long-lamented that “fiscal dominance” discourse often leads too immediately to “de-dollarization”, which we thought was simply incorrect. Lament no more. Relative to the major DMs (EZ, Japan, UK), the US is in far better shape, and the structure of the Eurodollar system is such that money may flow from EZ/Japan/UK to the US (we’ve explained and charted this precisely, FX-by-FX, in other publications). This realization seems to have dawned on meeting participants as a result of the obviousness of Europe’s, Japan’s, and the UK’s deeper problems. But also because the USD and the US yield curve have flattened in response to Fed tightening. The simple heuristic we’ve emphasized for the major economies is that if your bonds (prices) and currency are falling, your credibility is being questioned; the US isn’t in this box.
2. Europe is extremely vulnerable for well-known reasons. High debt in some, but the absence of fiscal and financial union means that the structure invites attack on individual sovereign bond markets such as France (your author’s birthplace) or Spain. The issue then becomes the likely political and policy response of the sovereign, which has to be highly uncertain in France, for example. France itself, not the EU or EZ, will have to fix France.
3. EMs remain largely exceptional, for durable reasons, but some are vulnerable — EM is not a monolith. EM exceptionalism is based on low government debt, less levered financial institutions, independent central banks, and often supported by strong commodity export revenues. It’s been our thesis. And it applies to both commodities exporters such as the higher-beta Latins, as well as to commodities importers such as lower-beta China and Thailand. China and Thailand are our poster-children for currencies that could become a normal part of a central bank’s reserve mix, sharing reserve status with the USD. Our point is that the exceptional EMs include defensive “safe” assets that are being demanded as reserve assets, challenging the usual prior that EMs are simply riskier and just pay a higher premium. We’ve noted in other publications that overall EM bond volatility is lower than DM bond volatility for several years already!
4. EMs are not all following the Fed. Mexico’s monetary policy had before this Fed hike cycle already diluted its heretofore usual principle to follow the US central bank’s hiking and cutting cycles. This can be ignored by markets in stable or declining rate environments, but will get an acute test in a rising US policy rate environment. To our eye, this points to a desire or willingness to let the peso weaken; that’s how it could end up looking to markets. Other EMs will also get tested, not necessarily on this precise dimension of the policy rate, but in terms of the monetary-fiscal system — India and Indonesia being obvious ones. We should also note that this divergence in many EMs is from a position of strength — Asian inflation has been lower than US inflation for several years already. China’s USD bonds trade through US treasuries. Many EMs are not going to hike along with the Fed, but the reasons are anchored inflation and inflation expectations, not worries over their financial systems’ abilities to handle higher rates. As we’ve charted in previous pieces, EM sovereign and financial vulnerabilities are of a different order due to low debt and leverage.

Where Is the Market in Denial? The Top 7 Blind-Spots and Their Asset Price Implications

Markets, like people, are often in denial. Mark Twain said “it is easier to fool someone than to convince them that they have been fooled”. We have had a front-row seat to this denial — virtually all G-10 economists saw “fiscal dominance” as an inappropriate frame for DMs. Many now acknowledge it, and the rise in gold prices forced many sell-side institutions to re-frame their thinking. But, EM economists have been given multiple observations of “fiscal dominance” playing out over time, unlike DM economists. DM economists have lived in an intellectual world where FX=1 (think about it), whereas EM economists know that FX ≠ 1. (Most G-10 economists, in particular, dismissed scenarios in which reserve currency countries could see monetary policy lose traction, the logical result of “fiscal dominance”, the precise output of which is both higher yields and a weaker currency. Furthermore, G-10 yield curve analysis, the term premium in particular, always refers to inflation, but not technical default, dependence on offshore borrowing, political crises, which are all real things to an EM economist.) Our point is that just because “fiscal dominance” now comes out of the mouths of DM economists, that is only the beginning of the end of denial, not universal acceptance. With that example and introduction, here are the top 7 market “denials”, to our eye:

1. DM political risk. Fiscal dominance always ends up being mediated by politics, and market participants still don’t anticipate how unpredictable these become in DMs, because the baseline expectation is political stability. We all know how unpopular Europe’s governments are. But even in the “exceptional” US, where we argue it is “least-bad” compared to other DMs, the K-shaped recovery is already becoming an intractable political challenge. This is an interesting example because the K-shape is partially driven by the US’s open capital account, consistent with its reserve status. The attractiveness of the US financial system and its assets is also driving the K. Asset conclusion: higher bond yields.
2. US exceptionalism. The realization that the US is “least-bad” among DMs is just dawning. In previous meetings, Europe was given the benefit of the doubt, and in some meetings there was great hope (remember Germany ending its debt-brake, the re-armament, the Draghi “suggestions” of fiscal largesse)? That is gone, and the EUR as we know is the only major fiat alternative to USD. In fact, some participants in Europe are for the first time wondering if they have vassal status, something very hard for them to understand or accept given their information diet. This seems to transcend the European anti-Trump consensus (among elites, not their officials), meaning Europe’s status in US thinking transcends US left/right politics. Asset conclusion: stronger US dollar.
3. Everyone remains an expert on geopolitics. Participants continue to largely ignore major geopolitical shifts. Part of this is defensible. Just as paying attention to US politics was fairly useless for predicting asset prices, geopolitical risk also always got absorbed, the oil always got out. Not now. The closing of the Gulf was the first in its history (over a thousand years), and Iran’s population size is regionally dominant. The Black Sea is under blockade by Russia, also population-dominant in the region. But, the core problem to our eye is deeper — most market participants today seem unaware of basic facts such as that there is always a “grown up” table and part of their agenda is propaganda (this is a normal perhaps necessary component of all conflicts). The Kursk offensive and all the “drones changed warfare” stories, for example, was a very successful operation in information terms, because it resulted in additional financing for Ukraine. But militarily it was not useful, the offensive was turned back and the Russian line moved forward. We’ve seen similar in the war in the Gulf. Asset conclusion: higher for longer oil.
4. China was everywhere and nowhere. Most importantly, participants did not have much exposure to China, that was the most striking observation, obviously bullish. Chinese officials rarely present at these meetings, but participants set an agenda that often includes China, but China itself was not a central discussion this time, to our way of looking. It was actually hilarious how China discussions were set up. Essentially, the US was offering generous advice to the Chinese, to revalue their currency due to “over capacity” or “subsidies and to then use their fiscal space to stimulate consumption. Because, you know, we care about the Chinese consumer, just look at how we’re treating the US consumer! Amazing that achieving low inflation is now a bad thing, but it makes sense if you are financialized as the US. Like I said, hilarious. It really speaks to China’s continued growth as a core “fact”. We are reacting to it, not it to us (and “we” don’t know this).

Now, we actually think all of this is going to happen — CNY will appreciate, fiscal will eventually get engaged, the economy will become more “balanced”; we wrote our predictions that CNY and CGBs would appreciate, after the April 2025 meetings. And, we appreciate the political pressure on China, but this is simply a tailwind to our view. China is going to decide what China is going to do, and the bottom line is — THE EXTERNAL SURPLUSES MAY CONTINUE INDEFINITELY. Asset conclusion: gradually stronger CNY and CGB prices.

5. India drifting, at best. India didn’t come up prominently at the meetings, which is a bit unusual given that it is now a significant portion of the GBIEM. A lot of this is because India is already fairly heterodox compared to other EMs (we’ve called it “not ready for primetime”), mostly centered on the RBI’s management of INR. A permanent shift up in commodities prices added to this general desire to “avoid” investing or even spending much time on it, on the part of participants. We are totally onboard with this attitude (INR has been a zero-weight in our strategies despite its large inclusion in benchmarks). One line we heard describing a prominent sell-side economist was “giving up on India”. We won’t get into the latest intervention in INR, which occurred during meetings; they are just the latest of a recent spate, and this spate is one of many, among the worst looks you can have. What we will observe, though, is that there seem to be no obvious bullish catalysts, and more than just catalysts might be needed if global commodity prices remain elevated. At the end of the day, real rates are just too low to provide a credible anchor and the exchange rate will remain under pressure. Asset conclusion: zero weight IGBs.
6. EM is not a monolith — there are winners as well as losers. We have details on many countries inside, so we’ll be very superficial in this paragraph. Winners are Brazil, Colombia, Malaysia, Thailand (a surprise to us), China, Pakistan, Peru, Uruguay, Paraguay, Kazakhstan, the Dominican Republic and many smaller commodities exporters, mostly in sub-saharan Africa. Losers are Indonesia, India, Philippines, Turkey, Mexico, Senegal, Ukraine, Romania, Lebanon, Mozambique. Eastern Europe is a potential loser due to China’s manufacturing challenge, energy supply/costs, and link to the EU’s issues. Asset conclusion: o/w winners, u/w losers.
7. AI is only about capex and productivity. This view, that one should only focus on the analyzable elements of AI, namely the impact of capex on growth, has been broadly held by participants, and we completely shared it. And, AI has been a key growth and, importantly, profit driver. This was the right lens. Similarly, this crowd (which includes us), is far more comfortable with a range of estimates or scenarios for the “productivity” gains from AI, and we emphasize how bad economists have been at this throughout history.

But this isn’t the right lens now. There are a multitude of new risks, the importance of which is magnified by AI debt issuance globally in 2026 transcending sovereign issuance (we’ve all also heard of AI’s growing dominance of IG issuance and benchmarks). Here’s our list, you’ve probably heard these elsewhere. Most precisely, rising rates will find problems, and these will be milestones, we’ve talked already for months about Oracle and CoreWeave being obvious lily pads on this path to discovery. Energy costs also strike us as under-appreciated risks, which obviously adds politics to risks. Profitability challenges at frontier labs from open-source models. Keep in mind, these ultimately have nothing to do with “AI’s future”, but the fact that AI is competing with sovereign debt issuance, in a rising rate environment means financing concerns are likely to dominate the discussion. Asset conclusion: UST prices, short term, depend on AI debt issuance.

What the IMF WEO and GFSR Said

Both publications are must-reads, in our view. They were released in Bangkok on October 13, in the middle of our meetings, and both are largely about shrinking buffers.

The key points from the WEO are:

• Steady headline, reshuffled interior. Global growth is projected at 3.1% in 2026 (3.5% in 2025) and 3.4% in 2027, while global headline inflation jumps to 4.7% this year before easing to 4.1% in 2027. The Fund’s own framing is that growth is being reallocated according to exposure to the war-induced energy supply shock and to AI-related demand. Economies riding the AI investment boom or favorable terms of trade are doing relatively well; energy importers and lower-income economies are not. We would simply add that this is the IMF’s way of saying EM is not a monolith.
• The baseline is a timetable, not a forecast of the war. The reference forecast assumes the Strait of Hormuz reopens in March 2027 and that prewar conditions are broadly restored by October 2027, with a petroleum price index averaging $96 a barrel in 2026 (16% above the April assumption) and natural gas about 36% above the April price. Policy and geopolitical uncertainty are assumed to stay elevated through 2027. What we heard in the meetings is that this assumption is what produces a V-shaped recovery for the Gulf: the Middle East and Central Asia region contracts 0.3% in 2026, with double-digit declines in Bahrain, Iraq, Kuwait and Qatar followed by double-digit rebounds in 2027. Given this assumption looks unlikely to be realized, 2027 growth and inflation will most likely be revised down (up).
• Advanced economies: 2.1% in 2026, 1.8% in 2027. The US is at 2.3% in both years (2027 raised), the UK at 1.3% and 1.2% (the Fund credits AI investment), the euro area at 1.1% and 1.2%, Germany at 1.3% and 1.1%, France at 0.4% and 0.7%, and Japan at 0.7% in both years. That is the “Europe is extremely vulnerable” point above in growth form.
• China at 4.5%, India at 7.0%. China is projected at 4.5% in 2026 and 4.0% in 2027, with second-quarter growth of 4.3% as weak consumption and property strains outweighed strong exports and higher oil prices weighed on activity. India is projected at 7.0% in both years, upgrades of 0.6 and 0.3 percentage points, on domestic demand, services and AI-related investment.
• Buffers are shrinking and the risks are to the downside. The Fund says the margins of adjustment that made the global economy resilient are narrowing rapidly. The risks it names are a persistent or escalating Middle East conflict that keeps energy and food prices high, rising sovereign bond yields, and renewed trade tensions. It calls the AI risk two-sided: expectations could prove too optimistic and trigger a market correction, or productivity gains could arrive sooner than expected.
• Fiscal space, again. Public debt is rising almost everywhere as yields climb, and the Fund has been saying that global public debt is approaching 100% of GDP. The Managing Director called advanced economies the worst offenders. The advice is to rebuild fiscal space without further delay, to keep any support for households facing higher energy and food prices temporary, targeted and non-distortionary, and to keep monetary policy data-dependent and tailored to each country. That is fiscal dominance, minus the phrase.
• EM and Southeast Asia. Fiscal policy in emerging and developing economies is expected to tighten gradually. Several Southeast Asian economies are projected to benefit from data-center activity and the global technology upturn, and in some cases from emergency fiscal measures, which fits our list of winners above. An accompanying IMF blog cautioned that the favorable window for EM may not last.

Source: International Monetary Fund (IMF), World Economic Outlook and Global Financial Stability Report, October 13, 2026. Projections are those of the IMF, are not guaranteed, and are subject to change.

The key points from the GFSR are:

• Resilience is not safety. The report’s first chapter is titled “Beneath the Resilience: Rising Risks, Unresolved Vulnerabilities.” Markets have held up on upward growth surprises, strong earnings and AI optimism, and financial conditions remain relatively accommodative, but the IMF says financial stability risks remain elevated.
• An AI-related correction is the first risk named. The Fund reviewed 14 circular-financing transactions since 2023 and found about $300 billion of equity investment and $650 billion of debt, in exchange for roughly $1.4 trillion of future revenue commitments. Hyperscalers are borrowing more and investors have leveraged into AI stocks. Stress at a key recipient would lower revenue forecasts and investment valuations, and weaken the value of the debt in those deals. It is the same financing lens we argued for above.
• Sovereign yields may be normalizing toward pre-GFC levels. Yields have risen on energy-driven inflation, fiscal concerns and, in some cases, stronger growth. The amplifiers the Fund flags are feedback loops between fiscal imbalances and leveraged bond trades, central bank balance sheet reduction, and a greater reliance on price-sensitive investors. Higher yields also raise discount rates, which depresses valuations elsewhere (AI-related assets included) and reduces the value of bonds used as collateral.
• Cyber is now a systemic risk. In more than 70% of reported cases this year, software providers had almost no time to patch a vulnerability before it was exploited, more than double the share in 2021, when providers typically had close to 300 days. Frontier AI models could accelerate both attack and defense.
• Banks are resilient overall, with a growing tail. Under the adverse scenario, banks holding 22% of global banking assets would end up below regulatory capital limits, up from 18% a year ago. Pension funds and insurers hold more illiquid assets and derivatives and lean on the same asset managers, private credit is carrying more leverage and on-demand funding commitments, and hedge funds now manage about $13 trillion, more than three times the 2013 level (the subject of an analytical chapter).
• EM: spreads are tight, yields are not. EM bond yields have risen while spreads over advanced-economy peers remain fairly compressed; the Fund describes this as a conundrum, we don’t see it that way because high growth and tight spreads seem synonymous to us. The IMF said it is particularly concerned about buffer depletion in some low-rated EMs and a growing reliance on less transparent funding sources in some frontier economies, and it urges EMs to rebuild buffers while financial conditions remain accommodative.
• The policy menu. Put debt on a sustainable path, limit leverage and liquidity mismatches in AI, private credit and other nonbank sectors (including sectoral capital and concentration requirements on banks’ exposures), and harden cyber defenses through better incident reporting and resilience testing.

Source: International Monetary Fund (IMF), World Economic Outlook and Global Financial Stability Report, October 13, 2026. Projections are those of the IMF, are not guaranteed, and are subject to change.

Our Country-Specific Comments

The notes below come from our own meetings and those of our colleagues during the week, which were held under the Chatham House Rule. They are organized by region and, within each region, by our view of how much the country matters to portfolios and benchmarks. Remarks are reported without attribution to any speaker or affiliation, and where a view comes from an outside participant rather than an official, we say only that.

Asia

China. A senior Chinese official skipped the usual state-of-the-economy talk and instead made the case for investing in RMB assets: lower volatility and lower correlation with global benchmarks than USTs or the S&P 500, tight bid-ask spreads, high turnover, a Panda market above CNY1.3 trillion and CGB inclusion in many benchmarks. The IMF has Chinese growth at 4.5% this year and 4.0% in 2027. One speaker argued the CNY should keep strengthening, perhaps to 6.6–6.5 per dollar by year-end, with capital outflows and a reverse Balassa-Samuelson effect from AI-driven services productivity as the limits; at least one large bank has also lowered its year-end USD/CNY target to 6.60. Another stressed the local-government problem (property-sale revenue down from about 8% to 2% of GDP, and VAT-sharing that rewards overcapacity) and said the authorities are comfortable absent an external shock. One participant dissented: no financial crisis, but the CNY is unlikely to appreciate, and bonds are the best asset class with rates heading lower, Japan-style. The same participant noted heavy Chinese imports of gold bars and coins, which he treats as savings rather than consumption.

Indonesia. A senior policymaker described the June–July hikes as frontloaded defense of the rupiah rather than a response to inflation, said policy is on hold, and walked through the toolkit meant to attract portfolio flows: relaxed hedging rules, a new FX-documentation framework, local-currency settlement arrangements with seven central banks, secondary-market IndoGB purchases (“not QE”) and SRBI. He was cautiously confident on the MSCI review, noting the authorities had gone above and beyond what MSCI asked. Later in the week MSCI said Indonesia no longer merits EM status and is weighing a “standalone” classification; the 10-year local yield jumped 10bp to 7.37%, and a final decision is due in November. Charm offensives sometimes fail.

Philippines. A senior policymaker said the focus is inflation rather than growth. 75bp of hikes have removed a lot of accommodation, a policy meeting is next week, and a severe El Niño scenario could push headline CPI above 15% year-on-year in Q1 2027. Infrastructure spending is down about 30% this year, with growth expected to recover in 2027. On the peso, it is the pace rather than the level that matters.

Latin America

Brazil. Brazil was on everybody’s mind. The 10-year yield sits around 13%, the runoff is close, the latest polls point to a tighter race, and betting markets have shaved the probability of a Flavio Bolsonaro presidency. President Lula is spending to win votes (free nationwide weekend transit, and the Senate just approved an end to the six-day work week), and undoing that largesse will be politically costly whoever wins. A change in political direction could also open the door to fiscal discipline and a rally in local rates.

Mexico. Policymakers are split much as they were in March and May: some weigh the strong peso and weak activity more, while the official we met weighs global risks, which look inflationary, more. Services inflation is showing incipient moderation, particularly in housing, non-core inflation is at atypically low levels (a reversal risk), and both the inflation gap and the ex-ante real-rate gap are close to zero. Fiscal measures have cushioned the energy shock, the economy is growing below potential but activity is rebounding on consumption and investment, and the central bank does not need to follow the Fed. There is a risk of Banxico embarking on a rate cutting cycle just as the Fed is doing the opposite. Two other points: transmission from policy to credit is weak (informality is above 50%), and the peso’s beta to the S&P and VIX is the fourth highest in EM.

Colombia. A senior policymaker told us that inflation is well above target and more persistent than in regional peers, with December 2026 CPI seen at 6.9% and 4.3% at end-2027, both including El Niño. Tightening was frontloaded, 12.75% may be the limit of this cycle, and a 10% minimum-wage increase, if it happens, would force more hikes. The structural fiscal deficit has been widening for ten years, and a credible plan is needed before the central bank can turn less hawkish. A sell-side political-risk assessment puts at 60% the chance that the new administration’s fiscal rescue law passes in watered-down form; the plan includes asset sales (the Ecopetrol stake among them), an end to fuel subsidies, a smaller state and a debt reprofiling to cut interest costs, and the government is eager for an IMF precautionary line or stand-by.

Argentina. The tone was optimistic. Officials argued that Argentina has changed culturally on fiscal discipline, pointed to structural reform passed with limited support in Congress (RIGI, labor and tax measures), and compared RIGI with Chile’s DL600 in 1974. Only $6 billion of $100 billion of approved mining investment has been made, so the pipeline is large. Officials also described a cleanup of the central bank’s balance sheet, a money-supply-based program with FX flexibility, and dollars slowly returning to the banking system. Fiscal credibility was described as the president’s most important asset, with the tax code and labor reform to come after the elections. The market has internalized past achievements; what investors now want is reassurance that the reforms survive into the next presidential term.

Peru. A senior policymaker said private employment is growing about 6%, the current account surplus is about 4% of GDP, the fiscal deficit is about 1.5% of GDP this year and next, and public debt is about 26% of GDP. Headline inflation of 4.5% is above target but is about 1.8% excluding energy, food and transport, and El Niño is at its peak. The message was that a hike is coming and the question is timing: one, at most two or three, and no desire to follow Brazil’s path. Politics are the drag (a long run of short-lived presidents), though a restored two-chamber congress should make a full five-year term likelier. Soberano yields are expected to decline over three to four years.

Chile. A senior policymaker said the central bank is moving with caution: oil pressures could prove persistent and the reconstruction bill’s tax cuts should lift investment from early 2027. Growth has been weaker than expected because of mining, unemployment is edging up, and the output gap is about -1%. The FX accumulation program is small (1–2% of daily turnover) and the policy rate is close to the middle of the neutral range.

Uruguay. The economy is more diversified and less exposed to the energy shock: oil imports are 1.6% of GDP versus nearly 7% in 2008, and fossil fuels’ share of power generation fell from 50% to 2%. The government targets a balanced primary result by 2030, with the headline deficit falling from 4.1% of GDP in 2026 to 2.5%, but most of the measures are revenue-based, which carries implementation risk. Reserves are 20% of GDP (1.96x the IMF’s adequacy metric) and officials see strong signals on OECD accession.

Dominican Republic. The economy is doing better than expected despite the energy shock. Inflation peaked at 5.7% in June and is 5.1%, with a return to target expected in Q4; growth is running about 4.5%; the current account deficit should be about 1% of GDP; and the IMF has reclassified the exchange-rate regime as floating. The budget deficit has been revised to 3.6–3.7% of GDP on energy subsidies. Officials do not expect the central bank to have to follow the Fed if the Fed hikes further.

Paraguay. Officials expect growth of 4.5% in 2026 and 4.2% in 2027, and public debt is about 38% of GDP. In 2026 the fiscal rule was relaxed, lifting the deficit target from 1.5% to 3.2% of GDP to clear arrears of about 2% of GDP in health and construction left by the previous administration, with the return to 1.5% pushed to 2028. The Atome project is on hold after the President revoked its decrees in June amid a power-pricing dispute with Brazil.

Emerging Europe and Central Asia

Turkey. A senior policymaker said disinflation is resuming in the categories most sensitive to monetary policy, while energy and the conflict remain the key risks and inflation expectations have not improved much. Reserves, adjusted for the decline in the gold price, are close to prewar levels. The faster pace of lira depreciation was described as a mechanical adjustment to keep monthly depreciation near 1.8%, not a change in policy and not a level target, and it is premature to talk about floating. The bigger risk is domestic dollarization rather than capital outflows, and there will be no easing and macroprudential tightening at the same time. Higher US rates would make policymakers more cautious on the pace of easing. The domestic fund crisis risks a downgrade by MSCI of Turkey’s emerging market status.

Hungary. A senior policymaker said monetary policy has to remain restrictive in the sense of keeping real rates positive. The government’s fiscal consolidation package is still undecided and is the key risk, there is no euro-adoption path, and most Maastricht criteria are not met. The current account is near balance and reserves are EUR62 billion versus EUR34 billion in February 2022. 2026 growth has been revised down to 1.8% on weak private investment, with companies waiting to see the 2027 budget; EU funds (RRF and SAFE loans) should arrive before year-end. The inflation target moves to 2.5% from January 2028.

Czech Republic. The assessment we heard is that the economy has surprised on resilience, particularly household consumption, and that fiscal policy is loose and can loosen further in 2027. Core inflation is stable but elevated and domestic demand is the main inflation pressure, so the official we met sees a possible need for additional tightening; the timing may be adjusted for the ECB but not the trajectory. Geopolitics has helped manufacturing, particularly defense.

Romania. Fiscal consolidation of about 3% of GDP over two years is underway and GDP is contracting in 2026, with a negative output gap of close to 4%, nominal wages down 3% and employment falling, so the labor market is no longer an inflation source. The leu is seen as about 1% overvalued versus 5% a few years ago. The EU’s RRF program has ended and is to be replaced in part by SAFE loans (a first tranche in November), FDI is falling, and the excessive-deficit procedure requires a 3% deficit within seven years.

Kazakhstan. Growth was 6.5% in 2025 and 4.1% so far this year, inflation has fallen to 9.3% from 13% a year ago, and policy rates have been cut by 175bp. Gold is 77–78% of reserves and officials are not comfortable with such a “mono-gold” composition; about 50% was described as the right long-run share and 75% as defensible for now given geopolitical risk. On local-currency index inclusion, officials had a positive conversation with the index provider; if the March–May 2027 client survey is positive, inclusion in GBI-EM could come as early as September 2027. (Kazakhstan is, incidentally, the one country in our gold analysis whose gold already covers more than 100% of M0 at current prices.)

Ukraine. Officials said forecasts will be revised significantly on intensified Russian shelling: 2026 growth, forecast at 1.8%, will be revised to slightly negative, and September inflation of 10% compared with a forecast of 8.6%. Rates have been hiked twice, from 15% to 16%. Alternative export routes can handle at most about 3 million tons, against the 4–5 million Ukraine normally ships, and FX intervention has risen to $5–6 billion a month, so FX liberalization is on hold. The summary of the situation was a war of attrition, with the hope that it is not Ukraine that blinks first.

Middle East and Africa

The Gulf. We heard that the baseline assumes the conflict ends by year-end and oil returns to prewar levels in 2027, which is what produces the V-shaped recovery, and that there is no known bailout plan for Bahrain. For Saudi Arabia, the message was resilience: Vision 2030 proceeds with full determination, buffers are huge and the non-oil sector remains robust. Across the region, tourism has fallen in single digits rather than the 30% or more first feared. Bahrain is expected to run a fiscal consolidation of the kind Oman did, with solid regional support. On Iran, the view was that economic pressure alone is unlikely to force a collapse, and that $100 oil would not cause one anywhere in the region.

South Africa. An official said growth is weaker and inflation higher, with gold a useful buffer against dearer fuel. The ex-ante real policy rate is only slightly above the 2.8–2.9% neutral estimate, so the main risk is finding itself insufficiently restrictive, and smaller hikes are preferred while conditions are monitored. Business inflation expectations respond more to direct engagement than to central bank signaling. Eskom’s reform has lifted potential growth from about 0% to about 1%, in part through confidence.

Egypt and Lebanon. Egypt has one IMF review left before year-end and no immediate financing need, so a new program is up to the government. Lebanon’s delegation was in Bangkok and the central issue remains a comprehensive solution for the banking sector, with no clear timeline. Restructuring advisers described the process as not advancing and reforms as not moving.

Senegal and sub-Saharan Africa. Growth in sub-Saharan Africa is seen at 4.5% in 2026 and about the same in 2027, increasingly uneven: oil and mineral exporters are improving while oil importers are slowing. Interest costs have risen to 13% of revenue on average, with some countries above 25%, and about one-third of countries are at high risk of debt distress or in it. Senegal reached staff-level agreement in early September and is the first test case for the G20’s sovereign-debt roundtable, but the creditor committee has not yet been formally constituted, treatment of domestic debt and the TRS is open, and board approval cannot proceed without enough progress with creditors (nor can a committee function without China). Congo’s program is tied to Gabon’s audit and the CEMAC peg question, and Mozambique’s FX market is not working. The revised LIC-DSF is approved but not operative until September 2027.

Some Observations on Thailand, Compared to My Childhood There in 1979

Your author lived in Thailand from 1975-1979, his parents (US Foreign Service) being sent there after the US evacuation of Saigon, which resulted in Thailand being the largest host of US forces outside the US at the time. My sister and I lived in an armed compound with mostly expats but also some Thais. Snakes and big lizards were a serious thing in our part of the city in those days, it’s hard to believe. We bought our house dog at Chatuchak market, but everyone seemed to end up with several strays, it was just the way it was. Singapore was home to our rival school, I saw a Singapore that is unrecognizable today. Anyway, it was hard not to notice a bunch of changes which resonate (positively or negatively) with many countries, not just Thailand.

• No strays! Didn’t see one stray dog. When I lived there, Bangkok was beginning to euthanize them, as they were a clear public health risk. Rabies treatments for humans were common.
• No police, no army, no guns! There were several coups or coup attempts when I was there, and perfectly seasonal rioting by students, normally violent. Once a year, every major intersection had not just police but military back up behind sandbags. Police were ubiquitous also for crime, not political unrest.
• No more khlongs! Bangkok traffic was never a problem because we always used khlong boats. They basically were becoming open sewage, so we’ll have to take their being filled to make roads as progress.
• No more billboards on the dangers of hard drugs. Not sure what to make of this, one could assume it is a sign that the problem isn’t as acute; I saw no evidence of this problem today, whereas it was visible then beyond the billboards. Related, because I’ve heard it described as a mild narcotic (like nicotine), but most older Thais in my day had red-stained teeth from chewing betel. I was a bit shocked, nonetheless, to learn that marijuana was somehow legalized, very unique in Asia.
• It remains by far my favorite country outside the US, for all the evident reasons. But the people really are special in meaningful ways. The service culture is an example, but it is deeper. My best example of Thai peaceability is the “bending with the wind” strategy when Japan in 1941 initially moved on Bangkok. The result of this was that the US ended up not declaring war on Thailand, making post-war arrangements far more successful. (I realize that there was a domestic political dynamic behind this, which is not what we are discussing). That remains, and is a core strength in a world with shifting alliances.

Fun Comparison of Thailand in 1979 and Today (with the US for Reference)

Source: World Bank and UN World Population Prospects, via GeoRank.org (retrieved 2026); “today” is 2025 for population, GDP and GDP per capita, with life expectancy and fertility the latest estimates. Rabies: Thai Department of Disease Control and published literature (370 deaths in 1980, the reported peak year, equal to 0.78 per 100,000; 7 deaths in 2025 as reported by the DDC); US CDC (22 cases in 1980–97; 17 cases in 2015–24). Clinical rabies is almost invariably fatal once symptoms appear, so reported cases and deaths are effectively the same count; what prevents cases is post-exposure vaccination after a bite. The last four rows are your author’s own observations.