FLOAT should not be used for Money

[Rick James]

The problem has to do with what FLOAT can and cannot do.

It cannot represent 1.1 exactly.

Also, FLOAT and DOUBLE are representations that have a different amount of precision. When you say
WHERE price = 1.1
and price is FLOAT, you are mixing precisions and/or FLOAT vs DECIMAL.

I have a table with
  `b` decimal(3,2) DEFAULT NULL,
  `f` float DEFAULT NULL,
  `d` double DEFAULT NULL,

INSERT INTO deci (b,f,d) VALUES (1.1, 1.1, 1.1);

SELECT b, f, d, b=f, f=d, b=d, b=1.1, f=1.1, d=1.1 FROM deci
+------+------+------+------+------+------+-------+-------+-------+
| b    | f    | d    | b=f  | f=d  | b=d  | b=1.1 | f=1.1 | d=1.1 |
+------+------+------+------+------+------+-------+-------+-------+
| 1.10 |  1.1 |  1.1 |    0 |    0 |    1 |     1 |     0 |     1 |
+------+------+------+------+------+------+-------+-------+-------+

In the last 6 columns, note that the result of an '=' test is either 0 (meaning FALSE) or 1 (meaning TRUE). The FLOAT version of "1.1" always fails to compare equal.

Bottom line: For Money, use something like DECIMAL(8,2).