For example, when a name refer to both a table column and a routine parameter. The execution of such code might lead to errors or unexpected results due to counter-intuitive resolution logic. Usually, names with a more local scope have higher priority.

Example (PostgreSQL):

CREATE TABLE foo
(
    id   INT,
    name VARCHAR(5)
);
CREATE FUNCTION func(name VARCHAR(5)) RETURNS INT AS
$$
DECLARE
    b INT;
BEGIN
      -- `name` is ambiguous as it is used as a column name and a parameter
    SELECT COUNT(*) INTO b FROM foo t WHERE t.name = name;
    RETURN b;
END;
$$ LANGUAGE plpgsql;

In PostgreSQL, you can use the #variable_conflict directives to explicitly specify a correct reference. For example, use #variable_conflict use_column to refer to a column name, or #variable_conflict use_variable to refer to a parameter.

CREATE TABLE foo
(
    id   INT,
    name VARCHAR(5)
);
CREATE FUNCTION func(name VARCHAR(5)) RETURNS INT AS
$$
    #variable_conflict use_column
DECLARE
    b INT;
BEGIN
    SELECT COUNT(*) INTO b FROM foo t WHERE t.name = name;
    RETURN b;
END;
$$ LANGUAGE plpgsql;