Okay, adding test data, fixing most of the syntax errors, breaking out the steps, and adding debugging statements (collectively referred to as "Troubleshooting by firepower" by an old friend in the business), we have some clues for you to examine:

#!/usr/bin/perl -w

use strict;

# Set up test data
my    @header =
(
    'Name',
    'DOB',
    'DOJ',
    'Country',
    'Region',
    'Salary',
    'Place',
);

my    @data =
(
    'JKSLK',
    19890101,
    20000101,
    'France',
    'ALMERIA',
    100000.00,
    'BCFGHK',
);

my    $empid    = 123456;


# Set up working data
my    @report = ();
my    @array = ();


# Initialize
push @array, @header;


# Do the work
my $csvfile;        # SKM:  What is this for, exactly?

my $reportLine = join "\t", ($empid, @array);
print "DEBUG:  \$reportLine = [$reportLine]\n";

push @report,$reportLine,"\n";
foreach my $reportElement (@report)
{
    print "DEBUG:  \$reportElement = [$reportElement]\n";
}

foreach  (@report) {
    print $_;
    print $csvfile $_;
}

exit;

__END__
[download]

This produces:

C:\Steve\Dev\PerlMonks\P-2013-07-11@1436-CSV-Fail>perl csvfail2.pl
DEBUG:  $reportLine = [123456   Name    DOB     DOJ     Country Region
+  Salary  Place]
DEBUG:  $reportElement = [123456        Name    DOB     DOJ     Countr
+y Region  Salary  Place]
DEBUG:  $reportElement = [
]
Can't use an undefined value as a symbol reference at csvfail2.pl line
+ 54.
123456  Name    DOB     DOJ     Country Region  Salary  Place
[download]

My observations:

1. The tabs are in the data. You already knew that, of course, since when you display it directly to the console you get the tabs. Your real issue is in how to get the information into a file, as you have stated. But there are other issues here which should be resolved first.
2. You don't seem to have the syntax figured out for writing to a file from Perl. Suggest you review the open, print, and close functions in the documentation, and look for code examples. It is rare to need to a scalar in place of a hard-coded file handle.
3. You are putting the newline on its own element in @report instead of simply appending it to the line. This is probably not what you intended, though it's not technically an error.
4. $empid is handled oddly here; I can see a number of reasons you are including it in the snippet, but without a functioning snippet, they are just guesses.
5. It looks like you are building a prototype of the logic to be used inside a loop. Not a bad approach, but coding up a fake loop is very little effort and in my humble opinion is better than writing linear code and then converting it later into a loop. Your mileage may vary.

I don't want to just solve the problem for you, since the point of PerlMonks is, generally, to help you become a better Perl programmer. That is best done if you do some of the analysis yourself.

I'd be happy to answer any questions you have about what this example is showing us, or adjustments to the code which more clearly refine the scope of your question.