Case supporting GNI as more useful. GNI measures income earned by a country's residents, whether that income is earned at home or abroad — i.e. GNI=GDP+net income from abroad. If the purpose of national income statistics is to track the material welfare of a country's citizens, GNI is the more direct measure: it captures remittances flowing in from citizens working overseas, dividends and interest earned by residents on foreign-held assets, and (via the negative component) income earned domestically but owned by foreign residents and sent abroad. For a country such as the Philippines, where several per cent of GDP equivalent flows in each year as remittances from overseas workers, GNI materially exceeds GDP and is a better reading of what the resident population actually earns.
Case supporting GDP as more useful. GDP measures output produced within the borders, which is the correct base for several practical purposes. First, it is the measure most closely tied to domestic employment and productive capacity: a country's workforce, factories and infrastructure produce GDP, not GNI, so GDP tracks the state of domestic industry and labour markets better. Second, GDP figures are typically published faster and with less revision than GNI, because net income from abroad relies on balance-of-payments income data that arrives with a lag. Third, GDP is less distorted by profit-shifting: in countries where multinationals book intellectual-property income (Ireland's well-known swings in GNI driven by foreign-owned IP) GNI can swing wildly year to year, while GDP (though also affected) is a somewhat more stable reading of real activity.
Reasoned judgement. Neither measure is universally more useful; the better choice depends on the question being asked. If the question is 'how well off are this country's citizens?' GNI is the right measure — welfare depends on income accruing to residents, not output produced on their soil. If the question is 'how is this country's domestic economy performing?' GDP is better — employment, productive capacity and tax base all track output within the borders. For most large economies where net income from abroad is small (under 1% of GDP for the United States, for instance), the two measures move together and the choice is immaterial. For small open economies with large cross-border labour flows (the Philippines, Nepal) or large foreign ownership of domestic capital (Ireland, Luxembourg), the gap is material and the measure matters; in those cases, triangulate between GDP and GNI rather than relying on one alone. The strongest formulation is therefore: GNI is the better welfare measure, GDP the better domestic-activity measure, and in small open economies both should be reported side by side.