{"id":18886,"date":"2026-08-19T12:05:08","date_gmt":"2026-08-19T04:05:08","guid":{"rendered":"https:\/\/maddruid.com\/?p=18886"},"modified":"2026-08-19T12:08:58","modified_gmt":"2026-08-19T04:08:58","slug":"3029-overemphasizing-ai-as-malaysias-second-quarter-2026-growth","status":"publish","type":"post","link":"https:\/\/maddruid.com\/?p=18886","title":{"rendered":"[3029] Overemphasizing AI in explaining Malaysia&#8217;s second quarter 2026 growth"},"content":{"rendered":"<p>Malaysia&#8217;s strong GDP growth in recent quarters has been attributed to sectors relating to artificial intelligence build-up. More often than not, that means semiconductor-making\/packaging\/assembly and data center construction. This brings up the question of whether this growth is standing on one leg?<\/p>\n<p>The rough answer I found is no.<\/p>\n<p>One quick and dirty way to answer that question is to strip those sectors out of the whole equation and have an adjusted GDP consisting of the rest of the economy. If that adjusted GDP growth is weak relative to full GDP numbers, that may suggest other sectors are not doing well and perhaps evidence of a one-legged economy.<\/p>\n<p><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-full wp-image-18887\" src=\"https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Full-adjusted-GDP.png?resize=580%2C374&#038;ssl=1\" alt=\"\" width=\"580\" height=\"374\" srcset=\"https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Full-adjusted-GDP.png?w=640&amp;ssl=1 640w, https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Full-adjusted-GDP.png?resize=300%2C194&amp;ssl=1 300w\" sizes=\"auto, (max-width: 580px) 100vw, 580px\" \/><\/p>\n<p>From the chart above that strips AI-related sectors, it shows the rest of the economy is doing well too since Malaysia\u2019s adjusted GDP growth would still be above 5%.<\/p>\n<p>The chart below shows the difference between the full and the adjusted GDP, which just reflects the earlier graph in a different way. Meaning, it presents more clearly how the AI-related sectors are becoming more and more important to overall growth, even as the overall economy has been expanding at a good rate.<\/p>\n<p><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-full wp-image-18888\" src=\"https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Full-adjusted-GDP-diff.png?resize=580%2C374&#038;ssl=1\" alt=\"\" width=\"580\" height=\"374\" srcset=\"https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Full-adjusted-GDP-diff.png?w=640&amp;ssl=1 640w, https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Full-adjusted-GDP-diff.png?resize=300%2C194&amp;ssl=1 300w\" sizes=\"auto, (max-width: 580px) 100vw, 580px\" \/><\/p>\n<p>It is possible to be more precise by using input-output table by taking into account things like water and electricity consumption specifically and more generally, all sectors affected by AI-related sectors (notwithstanding the latest official table is outdated by 5 years). But I would imagine as a back-of-the-envelope calculation, this method is good enough.<\/p>\n<p>If you do not like the approach, there is an easier but messier way to show this: the distribution of growth across all sectors.<\/p>\n<p><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-large wp-image-18889\" src=\"https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Sectoral-distribution-of-GDP-growth-2026Q2-Malaysia.png?resize=580%2C524&#038;ssl=1\" alt=\"\" width=\"580\" height=\"524\" srcset=\"https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Sectoral-distribution-of-GDP-growth-2026Q2-Malaysia.png?resize=1024%2C926&amp;ssl=1 1024w, https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Sectoral-distribution-of-GDP-growth-2026Q2-Malaysia.png?resize=300%2C271&amp;ssl=1 300w, https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Sectoral-distribution-of-GDP-growth-2026Q2-Malaysia.png?resize=768%2C694&amp;ssl=1 768w, https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Sectoral-distribution-of-GDP-growth-2026Q2-Malaysia.png?resize=1536%2C1389&amp;ssl=1 1536w, https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Sectoral-distribution-of-GDP-growth-2026Q2-Malaysia.png?resize=1200%2C1085&amp;ssl=1 1200w, https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Sectoral-distribution-of-GDP-growth-2026Q2-Malaysia.png?w=1583&amp;ssl=1 1583w\" sizes=\"auto, (max-width: 580px) 100vw, 580px\" \/><\/p>\n<p>The big chart above can be summarized:<\/p>\n<p><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-full wp-image-18890\" src=\"https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Share-of-sectoral-growth-2026Q2-Malaysia.png?resize=580%2C373&#038;ssl=1\" alt=\"\" width=\"580\" height=\"373\" srcset=\"https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Share-of-sectoral-growth-2026Q2-Malaysia.png?w=642&amp;ssl=1 642w, https:\/\/i0.wp.com\/maddruid.com\/wp\/wp-content\/uploads\/2026\/08\/20260819-Share-of-sectoral-growth-2026Q2-Malaysia.png?resize=300%2C193&amp;ssl=1 300w\" sizes=\"auto, (max-width: 580px) 100vw, 580px\" \/><\/p>\n<p>So in summary: yes, these AI sectors are positive for Malaysian growth but maybe we are are focused on the sexy parts of the economy while ignoring the robustness of a great many industries.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Malaysia&#8217;s strong GDP growth in recent quarters has been attributed to sectors relating to artificial intelligence build-up. More often than not, that means semiconductor-making\/packaging\/assembly and data center construction. This brings up the question of whether this growth is standing on one leg? The rough answer I found is no. One quick and dirty way to [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[10],"tags":[2193,2194,304,2147,2033],"class_list":["post-18886","post","type-post","status-publish","format-standard","hentry","category-economics","tag-artificial-intelligence","tag-data-centers","tag-gdp","tag-industries","tag-semiconductors"],"_links":{"self":[{"href":"https:\/\/maddruid.com\/index.php?rest_route=\/wp\/v2\/posts\/18886","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/maddruid.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/maddruid.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/maddruid.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/maddruid.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=18886"}],"version-history":[{"count":4,"href":"https:\/\/maddruid.com\/index.php?rest_route=\/wp\/v2\/posts\/18886\/revisions"}],"predecessor-version":[{"id":18894,"href":"https:\/\/maddruid.com\/index.php?rest_route=\/wp\/v2\/posts\/18886\/revisions\/18894"}],"wp:attachment":[{"href":"https:\/\/maddruid.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=18886"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/maddruid.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=18886"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/maddruid.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=18886"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}