I second others' comments that acknowledgements are low-cost and the indirect support of a scholarship in allowing one to engage in research may be sufficient reason to include it.
But here's another way of thinking about it.
Many agencies you will get funding from for your academic work are only the lowest members of a towering hierarchy of agencies. As an example, consider the US Department of Energy's Org Chart:
Say I get a small grant from the Western Area Power Administration, which is nested under the Assistant Secretary for Electricity under the Office of the Under Secretary for Science and Energy under the Office of the Secretary under something... something... the POTUS.
Each level within an organization can compete for funding with other agencies at the level in a zero-sum fashion. The output of fundees is evidence that an organization is using its funding well and should continue to receive that funding, rather than it being given to one of many other needy, competing agencies.
This is one reason agencies like XSEDE supercomputing place such a strong emphasis on citations and standardized acknowledgements: this allows them to mine publications to justify their continued budgetary allocations.
Because acknowledgements are helpful to my funders and I want to reciprocate the help they've given me, I try to build a "big tent" with my acknowledgements and err on the side of including folks.