The Two-Thousand-Ringgit Floor

Friday afternoon, 9 October 2026, the Dewan Rakyat.

Anwar Ibrahim, as Prime Minister and Finance Minister, read paragraph 32 of the budget speech. The heading of the section: "Upah Bermaruah untuk Rakyat" — a dignified wage for the people.

"Oleh itu, Kerajaan bersetuju untuk meningkatkan kadar gaji minimum daripada 1,700 kepada 2,000 ringgit." The government agreed to raise the minimum wage from RM1,700 to RM2,000, effective June 2027. ^1,2

After the applause, he read the next sentence: micro, small and medium enterprises (MSMEs) with annual revenue below RM50 million are exempt, "bagi memberi ruang untuk penyesuaian model perniagaan" — to give them room to adjust their business models. He said this means more than 4 million workers will be covered by the new minimum wage. ^1,3

In the same paragraph, the speech itself used a metaphor: the minimum wage "hanya menaikkan lantai pendapatan pekerja" — only raises the floor of workers' income.

A floor, not a building#

The metaphor of the floor is apt. It does not care how tall the building is or who lives in the penthouse. It guards one thing only: below which level a person must not fall.

So this piece asks three questions. How was this floor laid, board by board? What happened to the people inside the building when other countries raised their floors? And why raise it now?

You will find that the answer to the third question is not in the economics textbooks. It is in the political calendar.

Thirteen years, seven raises#

Malaysia came late to a national minimum wage. In 2011, Parliament passed the National Wages Consultative Council Act (Act 732); in July 2012, the first minimum wage order was gazetted. ^4,5

From 1 January 2013, the monthly minimum wage was RM900 in Peninsular Malaysia and RM800 in Sabah, Sarawak and Labuan. Small employers with five or fewer workers came under the rule half a year later. ^5,6

The path since then runs roughly as follows:

^4,7,8,9,6

One move recurs in this table: each time the floor is raised, the government leaves small employers a buffer. In 2013, 2022 and 2025, the buffer was drawn by headcount, and it had an end point — a few months later, the small employers had to catch up too. ^7,5

What is different this time#

In 2027, the buffer has changed shape.

The speech draws the line by "revenue", not by headcount; the word it uses is "dikecualikan" — exempted, not deferred. The speech does not say until when the exemption lasts. ^1,2

This is no small detail. The Malaysian Employers Federation (MEF) welcomed the arrangement, but also asked the government to spell out how the RM50 million threshold is calculated. ^10,11 Another employer group, Micsea, according to Free Malaysia Today, proposed transition support for firms with revenue between RM50 million and RM100 million, and said: "The aim is not to delay compliance, but to keep the transition from becoming a punishment."

On the other side stands the Malaysian Trades Union Congress (MTUC). According to The Star, it originally demanded RM3,100, considers RM2,000 not enough, and also opposes exemptions by company type — the same job could pay a different wage depending on who the boss is.

Paragraph 33 of the speech carries a companion measure that few have reported: beyond MSMEs, a company that wants a tax deduction on wage expenses must pay wages through a bank account. According to the text of the speech, this is meant to close the room for hiring undocumented foreign workers.

How high is the floor#

Whether a floor is high depends on how far it sits from the ceiling. Economists usually measure this by dividing the minimum wage by the median wage — a ratio known as the Kaitz index.

The Department of Statistics' Salaries & Wages Survey Report 2025, released in September 2026, shows that the median monthly salary of Malaysian-citizen employees in 2025 was RM2,940, 5.3% higher than the year before.

Calculating from that figure ourselves: the current RM1,700 is about 58% of the median, and RM2,000 about 68%.

Two limits need stating here. First, RM2,000 only takes effect in June 2027; by then the median will most likely have risen, so the ratio will be lower than 68%. Second, this median is not the "full-time workers" basis the OECD commonly uses, so it cannot be compared line by line with other countries' official figures. 68% is an upper bound, not a forecast.

Next, who is standing near the floor. The Department of Statistics' formal-sector wage statistics show that in March 2026, employees earning below RM1,700 a month made up 7.9%. In a column in The Star, think-tank scholar Lee Heng Guie estimated from the Department's data that another 1.27 million people fall between RM1,700 and RM1,999 — a figure this site could not check directly on the Department's page.

There is one more yardstick: the living wage. In its box article "The Living Wage: Beyond Making Ends Meet" in the 2017 Annual Report, Bank Negara Malaysia estimated that, at 2016 prices, a single adult in Kuala Lumpur needed about RM2,700 a month to live decently.

This figure is often cited, but it has three boundaries: it covers Kuala Lumpur only, it is based on prices from ten years ago, and it is a provisional estimate. The central bank itself wrote that a living wage is generally not a statutory standard, and is a separate matter from the minimum wage. So all one can say is that RM2,000 still falls below the central bank's provisional estimate; one cannot say "two thousand is not enough to live on".

As for prices, the Department of Statistics' official inflation is in fact not high: 1.4% for the whole of 2025, and 1.9% in August 2026. The budget speech's argument is also not that inflation is out of control, but something else: the speech says incomes have not kept up with prices, and compensation of employees is only 33.9% of gross domestic product (the speech does not say which year's figure).

A question the world has argued over for thirty years#

Does raising the floor destroy jobs? This is one of the longest-running quarrels in economics.

In 1992, the US state of New Jersey raised its hourly minimum wage from US$4.25 to US$5.05, while neighbouring Pennsylvania did not move. Economists David Card and Alan Krueger telephoned 410 fast-food restaurants in the two states and found that employment in New Jersey had, relatively, increased. ^12,13

A few years later, David Neumark and William Wascher redid the same exercise using restaurants' payroll records and reached the opposite conclusion: employment in New Jersey fell by 4.6% in relative terms. ^13,12

The same raise, two sets of data, two directions. The dispute taught later researchers one thing: the conclusion often depends on how you measure.

Two ways to read it#

Thirty years on, researchers used much larger samples.

One group analysed 138 state-level minimum wage increases in the United States between 1979 and 2016 and found that five years after a raise, the total number of low-wage jobs was essentially unchanged; within the range where the minimum wage is up to about 55% of the median, no relationship between employment and the ratio could be seen. ^14,15

But another group, Neumark and Peter Shirley, went through US studies from the same period one by one and found that about eight in ten estimates were negative, especially for teenagers and the less educated. ^16,13

Neither side fabricated anything. The difference is in whom they look at: the former looks at all low-wage jobs, the latter gives more weight to the group directly affected. An independent review commissioned by the British government in 2019 offered a middle summary: in developed countries, the employment effect is close to zero and wages rise noticeably; prices are an important channel of adjustment; and what happens when the minimum wage is set very high, the evidence is "still accumulating". ^17,16

The price link shows up clearly in restaurant-industry data from Canada and the United States: when wage costs rise, restaurant prices follow, concentrated around the time the law takes effect. ^18,17

Even a single city can yield two conclusions. After Seattle raised its hourly minimum wage to US$13, a University of Washington study found hours worked in low-wage jobs fell by 6% to 7%; a University of California, Berkeley study that looked only at the restaurant industry found the number of jobs was unaffected. ^19,20 The two studies measure different things, so they do not refute each other.

Seven countries, seven endings#

United Kingdom: a national minimum wage was introduced in 1999, and the Low Pay Commission has recommended the annual adjustment since. The Commission says that over two decades it commissioned more than thirty studies and found no strong evidence of negative effects; in 2019 the minimum wage was about 59.8% of the median. ^21,17 This is the implementing body's self-assessment, so it should be read alongside independent research.

Germany: it had no national minimum wage until 2015. Scholars tracking employee records found that wages rose, employment did not fall, and low-wage workers moved from small, inefficient firms to large, efficient ones. ^22,23 A 2026 opinion paper from the Institute for Employment Research (IAB) says fears of mass unemployment "have so far not materialised", and only micro-firms and mini-jobs saw adjustment; but the 2027 minimum wage will reach 60.4% of the median, and it asks for an evaluation first. ^23,22

South Korea: between 2017 and 2019, the minimum wage rose sharply from 53% to 63% of the median. A paper on manufacturing firms with ten or more employees found employment fell by about 3%, but productivity and wages of the workers who stayed both rose; by 2021, the unions said the wage promise had not been kept and business said it exceeded what SMEs could bear, so both sides were unhappy at once. ^24,25

Hong Kong: a statutory minimum wage was introduced in 2011, with the first rate set at 47.9% of the median and about 11% of employees affected. ^26,27 Four years later the government told the Legislative Council that total employment was 250,000 higher than before implementation — but the economy was strong in those years, so this increase cannot all be credited to the minimum wage. ^27,26

Thailand: in 2012 to 2013, the daily wage was pushed sharply up to 300 baht. Research found employment did not fall, but also found that about four in ten daily-wage observations were still below the legal minimum — not that the raise did no harm, but that many people never received it. ^28,29

Indonesia: in 2001, the power to set the minimum wage was devolved to the provincial level. Research found that a rising ratio came with job losses in the formal sector, but the informal sector absorbed more, so net employment actually rose. ^30,31 An earlier study found the harm concentrated in small firms, with large firms even increasing hiring. ^31,30 What absorbed the shock was unprotected work — this is not good news.

Brazil: between 1996 and 2018, the real minimum wage rose 128%, explaining about 45% of the fall in income inequality; the employment shock was cushioned by workers moving to more productive firms. ^32,22

Putting the seven countries side by side, a few conditions can be drawn out: how far the floor is from the median, how fast it is raised, whether firms saw it coming, how tight the labour market is, how many people work outside the formal sector, how small businesses make the transition, and whether anyone can call a halt when job losses appear. ^17,23,24

These conditions are exactly the exam paper Malaysia now has to sit.

Why now: Anwar's calendar#

Back to the speech of 9 October. The minimum wage is only one item in it.

In the same budget, cash assistance (STR and SARA) totals RM16 billion, up from RM15 billion the year before; adults aged 18 and above who are not STR recipients get RM100 each in two payments, up to 13 million people; about 1.3 million civil servants receive a one-off RM1,500. ^1,33 The speech calls this an "expansionary" fiscal policy, with a 2027 deficit target of 3.3%, while for 2026 it is revised from the original 3.5% to 3.6%. ^1,33

The Edge's headline put it this way: after years of fiscal consolidation, Anwar loosens the purse strings.

Now the calendar. This Parliament's term expires in December 2027, and a general election must be held by February 2028 at the latest. ^33,34 The minimum wage takes effect in June 2027, and the second SARA payment of RM100 is set around the 70th National Day — two milestones, both falling before the latest possible polling day. ^1,34

The unity government is not calm inside either. In the July 2026 Johor state election, Barisan Nasional took 48 of 56 seats, leaving Pakatan Harapan with 8 — one partner in the federal government thrashing the other at state level. ^35,36

At the UMNO general assembly in September, Ahmad Zahid Hamidi, who is also Deputy Prime Minister, without naming anyone shouted "Enough is enough, guru!" — the "guru" pointing to the title he used for Anwar in a recording leaked in 2021. ^37,38 In the same speech he criticised the flip-flop on the BUDI95 petrol quota policy, and said UMNO was ready for the 16th general election. ^37,39

Criticism also came from within the government camp. Former Finance Minister Lim Guan Eng, the day before the budget, said it was the "last budget" to let people feel the benefits before they vote. ^34,33

Two Akmals, one question#

There are two Akmals in this round; do not mix them up.

The first is Economy Minister Akmal Nasrullah Mohd Nasir. Two weeks before the budget, he said whether the wage would rise to RM2,000 would be announced by the Prime Minister in the budget, and that the government had to balance workers' welfare against the capacity of MSMEs. ^40,41

The second is UMNO Youth chief Akmal Saleh. In January this year he urged UMNO not to be afraid of being the opposition, and in September he said UMNO Youth did not want to work with Pakatan Harapan. ^42,43

On the evening after the budget, according to Sinar Harian, Akmal Saleh issued six proposals, with a headline quoting him: "Gaji naik RM300, siapa akan bayar" — the wage rises by RM300, who will pay?

The report quoted his original words: "Yang dapat kenaikan gaji hanya sebahagian pekerja, tetapi yang mungkin terpaksa menanggung kenaikan harga barang ialah seluruh rakyat." — Only some workers get the raise, but those who may have to bear the rise in prices are the entire people.

His six points include: regular price monitoring, strict action against profiteering and price manipulation, temporary incentives and automation support for affected micro and small enterprises, phased implementation based on state and industry data, care for workers who get no raise, and regular transparency reports on the policy's impact.

Worth noting, the tone of these proposals is not opposition to a raise, but a question of who bears the cost — which happens to be the very "price channel" that the economics literature above has quarrelled over for thirty years.

For completeness, the status of his case: in August 2025 he was charged under Section 505(b) of the Penal Code over a Facebook post, pleaded not guilty, and was granted bail. ^44,45 He also applied to the High Court for judicial review to challenge the prosecution; the hearing was originally set for 16 June 2026, and the outcome is not found in the public record. ^46,47 The case has not concluded.

Limits of the evidence#

• Akmal Saleh's six proposals: this site read the original text in only one outlet, Sinar Harian, so it is treated as a single source, and the text is cited as "according to the report".
• The exact scope of the exemption: the new minimum wage order has not been gazetted; how "revenue" is calculated, and whether the exemption has an end point, is not found in the public record.
• Gazette numbers and texts of past orders: the Human Resources Ministry's gazette documents could not be opened on the day this site checked; historical rates rely on law-firm compilations and news reports.
• Malaysia's own empirical research: this site did not read the full text of any paper meeting an academic threshold that directly answers "did the 2013 minimum wage harm employment". The conclusions from the countries above cannot be applied directly to Malaysia.
• Several countries' figures rest on only one or two studies each: for South Korea, Thailand, Indonesia and Brazil, this site read only the paper abstracts or working-paper versions.
• The 68% ratio: calculated by this site from the Department of Statistics' 2025 median; it is not a figure published by any institution.
• Official statistics have a single origin: the median wage and the share below RM1,700 come from the Department of Statistics; the living-wage estimate comes from Bank Negara Malaysia. The news and columns that repeat these figures all trace back to the same documents and do not count as a second independent source, so the text names the source directly and attaches no citation marker.

---

In June 2027, the new minimum wage order will appear in the gazette, and only then will we know how the line is drawn and for how long.

Before that day, Anwar still has to get past the Melaka state election, Zahid is still waiting for an earlier general election, and Akmal Saleh is still asking the question: that RM300 added on, who exactly pays for it?

Note: At the time of writing (October 2026), the new minimum wage order has not been gazetted, and neither Akmal Saleh's criminal case nor his judicial review has a final outcome on record. This site will keep watching developments and update.