Look at these numbers:

$10
$100
$1,000
$1,000,000
$1,000,000,000
$1,000,000,000,000


Now consider these corresponding words:

ten dollars
hundred dollars
thousand dollars
million dollars
billion dollars
trillion dollars


Can you see the problem? The discrepancy? To the casual or untrained reader or listener, the ratios between between a million, a billion, and a trillion dollars do not seem intuitively different from the ratios between ten, a hundred, and a thousand dollars, when seen or heard in words.

Seeing them written out in figures does make a difference, although helping the average person understand the comparative magnitude of a trillion dollars as opposed to a billion dollars still can be difficult. But I suspect that the real problem begins with how we label (in English, anyway) the numbers that we use for counting.

Here is another list:

*one
*ten
*hundred
*thousand
ten thousand
hundred thousand
*million
ten million
hundred million
*billion
ten billion
hundred billion
*trillion

For the first four numbers above, we learn and read distinct words (marked *). But then as we continue the sequence, words are reused, and only every third word we encounter is new. This misleads us, unfortunately, into subconsciously assuming that the numerical distance between million, billion, and trillion is equivalent to the distance between ten, hundred, and thousand.

If we had unique words for the other steps (and those beyond them) on the list above - if ten thousand was called, say, decand, and hundred thousand was called cenand, and if ten million was called medillion, and hundred million was called mecillion, and so on - it might make understanding of really big numbers more intuitive.

Anybody want to take on this project and see it through to conclusion? :)