The department identified 598 000 instances of duplicate IDs. It has now resolved 412 096 cases where one person had multiple ID numbers and 20 971 cases where multiple persons shared one ID number.Applying new identity management techniques to existing systems and databases can yield impressive results leading to a significant return on investment.
It's also interesting to compare the number of "extra" ID numbers generated to the number of people that were issued someone else's number in error.
There were more than 28 instances of extra ID numbers issued to people for every instance of assigning the same ID number to multiple people.
There are many reasons for this including simple error. If two people have the same ID number, someone is likely to become inconvenienced and bring the matter to the attention of the authorities. If someone loses their child's documents, they may just start the ID process over again, getting a new number and the old number never gets used. So innocent errors of one type (people sharing an ID number) are far more likely to be corrected than the other type (more numbers than people).
Another reason for the difference in the quantity of each type of error, however, is fraud. The ghost worker fraud depends upon creating extra ID's for people that don't exist and then either hiring them to work for some publicly funded entity or enrolling them to collect some sort of social welfare benefit.
I suspect that the bigger part of the difference in observed database error rates is fraud.
