A three-decade study of Japan’s flattening Phillips Curve offers a sobering mirror for Bangladesh, where high inflation and rising joblessness now sit together the very combination economists fear most, writes Abdullah Shahid
A WALK through Tokyo’s Akihabara on any weekday afternoon will bring up something quiet, almost unsettling, beneath the neon glare: prices on coffee, ramen, manga and even electronics have barely budged in years. A walk through Dhaka’s New Market the same afternoon brings up an inverse picture. Vegetable prices climb between morning and evening. Rice prices have not stopped rising for three years. A graduate engineer behind a tea-stall has become a stock character of post-uprising Bangladesh.
These two scenes seem to belong to different planets. They are, in fact, two faces of the same question: what happens when the most reliable engine in macroeconomics the trade-off between inflation and unemployment begins to misfire?
That engine has a name. It is called the Phillips curve, after the New Zealand-born economist AW Phillips, who in 1958 first documented that countries with low unemployment tended to register higher inflation and the other way around. The intuition is straightforward. When jobs are plentiful, workers earn more, spend more and prices rise. When jobs evaporate, demand cools and prices flatten. For nearly half a century, central banks navigated their economies by leaning on this trade-off. They could pick: a hotter labour market with some price pressure, or quieter prices with some idle hands.
The trouble is that the engine has begun to sputter in different ways for different countries. Japan and Bangladesh are now the two clearest illustrations of that breakdown, sitting at opposite ends of the same broken curve.
Japanese mirror
I RECENTLY conducted an empirical study of Japan’s Phillips curve from 1988 to 2024, fitting an ordinary least squares model to 37 years of consumer price, unemployment and Brent crude data drawn from the World Bank and the FRED database at the St Louis Federal Reserve. The exercise was influenced by academic curiosity. Japan is the rich world’s most-studied warning. Its boom of the 1980s, when the land beneath the Imperial Palace was famously valued at more than all of California, ended in the implosion of an asset-price bubble in the early 1990s. The decade that followed earned an unflattering name: the lost decade. Inflation flatlined. Prices, in many years, fell.
The numbers tell their own story. Over the full 37 years, average inflation in Japan was 0.687 per cent. Average unemployment was 3.548 per cent. Both figures would be the envy of nearly any finance minister in the developing world. Yet, they describe a country that has, by its own admission, been struggling for three decades to convince its people that prices will rise again.
The model results were unsparing. A one-percentage point rise in unemployment was associated with a 0.877-point fall in inflation in the simplest specification. When oil prices and the last year’s inflation were added capturing supply shocks and the role of human expectations respectively the unemployment coefficient remained robust and statistically significant, but the explanatory power of the model rose only to an R-squared of 0.585. The Phillips curve in Japan exists, but it is a remarkably thin one. As the economists Fumio Hayashi and Edward Prescott concluded, the boom that built Sony, Toyota and Nintendo did not merely pause; it gave way to a structural decline in productivity growth. Paul Krugman, writing in 1998, called the situation a liquidity trap: a country where the central bank could lower interest rates to zero and the public would still refuse to spend.
The deeper takeaway is one the Bangladeshi reader should sit with. Japan did not slip into stagnation because of a single shock. It slipped because the expectations of ordinary Japanese people changed. Once a population stops believing that prices will rise, businesses stop raising them, workers stop demanding raises and the entire machinery of monetary policy loses traction. Expectations, once anchored to deflation, are nearly impossible to dislodge.
Bangladesh in mirror
NOW, consider Bangladesh. The official unemployment rate ticked up to 4.7 per cent in 2024 from 4.2 per cent in 2023, according to the Bureau of Statistics. That figure obscures more than it reveals. The unemployment rate among university graduates stood at 13.54 per cent, as reported in the General Economic Division’s State of the Economy 2025. Over one lakh apparel workers have lost their job in the year following the 2024 uprising, with the Asia Floor Wage Alliance documenting at least 258 factory closures. Around 84 per cent of the jobs sit in the informal sector, where distress is harder to count and easier to deny.
Inflation, meanwhile, has refused to behave. After peaking at 11.66 per cent in July 2024 amid the supply disruptions that followed the uprising, the headline number drifted down to 8.17 per cent by October 2025, the lowest reading in 39 months, before reversing course. By April 2026, the year-on-year rate was 9.04 per cent, with food, housing and transport leading the climb. Bangladesh ended 2025 with the highest inflation rate in South Asia, well above Sri Lanka’s 2.1 per cent, Pakistan’s 5.6 per cent and India’s sub-one-per-cent reading.
If the Japanese pathology is a Phillips curve gone flat, the Bangladeshi pathology is a Phillips curve gone missing. Inflation and unemployment are rising together. Economists call the condition stagflation and it was last seen in a serious form during the global oil shocks of the 1970s. The central bank’s policy rate of 10 per cent, the textbook response to inflation, has compressed private credit growth to around 6 per cent without bringing prices to heel because the inflation in question is not, in the main, a story of an overheated labour market. It is a story of cartelised supply chains, of currency pass-through from a taka that has slipped against the dollar, of administered prices and energy costs.
What Japan can teach Dhaka
THE temptation, when looking across at Japan, is to draw the wrong lesson. Japan’s affliction is deflation; Bangladesh’s is inflation; surely the two have nothing to say to one another. But the underlying mechanic the one the regression attempts to isolate is the same in both countries. Both are economies in which the simple textbook relationship between jobs and prices has weakened because something larger has shifted underneath.
In Japan, what shifted was expectations. Three decades of falling prices taught a nation to stop expecting any. In Bangladesh, what is shifting is the credibility of the institutions tasked with managing prices and jobs. When citizens cease to believe that the central bank can tame inflation or that the labour market can absorb its own university graduates, expectations adjust and policy loses its grip.
Three concrete lessons follow. First, monetary policy alone will not be enough. Japan tried zero interest rates, negative interest rates, quantitative and qualitative easing and forward guidance, and still failed to dislodge a deflationary mindset for the better part of two decades. Bangladesh, applying its own monetary medicine in the opposite direction, is now discovering the same limitation in reverse. Second, the official unemployment figure can mislead. Japanese firms practise labour hoarding, holding on to workers in downturns and, thereby, masking the depth of slack in the economy. Bangladesh has an informal-sector cousin of the same problem: with 84 per cent of jobs informal, disguised unemployment in agriculture and small commerce hide the true distress beneath a deceptively orderly headline number. Third, expectations must be earned, not announced. The Bank of Japan eventually managed to nudge inflation expectations upwards only after coordinated, credible action across fiscal, monetary and structural reform lines, and even then progress has been halting. Bangladesh Bank cannot tame inflation alone if syndicate pricing, exchange-rate volatility and fiscal disarray are left unaddressed.
New normal or fork in road?
THE conclusion of the Japan study is, I admit, blunt. The roaring momentum of the 1980s has subsided. The boom is over. Japan has settled into a low-growth normal and the world’s other ageing, low-productivity economies are increasingly settling in beside it.
Bangladesh is not Japan. It still has the demographics that Japan would pay a great deal to possess: a young population, a willing work force, a manufacturing base and an outsized remittance flow. But, Bangladesh is now at a fork. One road leads back to the trade-offs the Phillips curve once described where careful policy could choose between price stability and employment. The other road, the one the economy now appears to be travelling, leads to the no man’s land of stagflation, where neither lever pulls and credibility erodes by the month.
Japan reached its own fork in the early 1990s and chose, mostly by accident, the road of low expectations. The cost is now plain to see in the price of a coffee in Akihabara, frozen for 30 years. Bangladesh has its choice ahead of it. Whether the next budget, the next monetary policy statement and the next round of structural reforms are equal to that choice is, in the end, the question that should occupy the finance ministry, the central bank and all.
Abdullah Shahid is a final-semester undergraduate at the University of Massachusetts Lowell, completing a Bachelor of Science in quantitative economics and econometrics.
